The Guardian says ABP avoided millions in UK tax – that’s probably wrong.
One of the frustrating things about the media is their unwillingness to accurately call out very aggressive tax avoidance by individuals1No, nobody in particular comes to mind, contrasting with a completely slapdash approach to accusing companies of tax avoidance.
Exhibit A, courtesy of the Guardian:
The ABP Food Group are accused of tax avoidance by having a £63m 5% loan into a UK company, ABP UK, from a Dutch company, Trojaan. But Trojaan is funded by 0% loans from other companies (including in Jersey). So, says the Guardian and their experts, this is “aggressive tax avoidance” to reduce ABP UK’s tax bill.
Does it smell like avoidance?
We can run a quick sense check. What’s the value of this “aggressive tax avoidance”? £63m x 5% x 19% (tax rate) = £600k. ABP is a €4bn global company. Why would it avoid £600k of tax? That’s a drop in the ocean of its UK profits. So on its face, the story doesn’t really make sense. You don’t need any tax expertise to work that out, just basic math.2insert tired joke about arts graduates working at the Guardian
Would the avoidance actually work?
It’s not obvious it would. The UK has had rules for over a decade now which prevent a business magicking debt into the UK if the group doesn’t actually have external debt. The current iteration of this is the “corporate interest restriction“. This (very broadly3really I can’t stress enough how complicated these rules are, and how over-simplified a summary that is) caps interest deductibility at the lower of (1) 30% of UK EBITDA, and (2) the amount of external debt the group has worldwide.
So if all that’s going on is the stuff in the Guardian article – the group has no external debt, but lends £63m into the UK at 5% interest, then the corporate interest restriction will kibosh any relief on that interest.
Of course, the Guardian may be missing the full picture, and the group may have plenty of external debt, in which case this rule may not apply (but the 30% EBITDA cap still would). But that’s okay – if the group as a whole has external debt, then passing some of that debt to ABP UK via a series of intra-group loans is perfectly natural, and not tax avoidance at all. (And I’m guessing that’s what’s actually happening)
What about the ABP Dutch company that has €118m profit but pays only €1.1m in tax? Could be Dutch tax avoidance, but more likely this is a transfer pricing adjustment, and reflects another adjustment being made elsewhere (e.g. Ireland) – so the group is “flat” overall. Again – sense check – we shouldn’t expect large taxable profits in a Dutch company that’s just man-in-the-middle of a bunch of intra-group lending.
So there is no particular reason to think UK tax avoidance is going on – the numbers are too small, and the “scheme” is one that’s countered by legislation.
Why did the Guardian think otherwise?
Because they didn’t speak to anyone with knowledge of UK tax avoidance – they quote a US tax professor and a UK campaigner with no tax expertise. That’s disappointing – there are literally thousands of people who know about this stuff.
Are you sure there’s no avoidance going on?
No, which is why I say the story is “probably” wrong.
A smart person on Twitter suggested a possible “double dip” structure, with ABP getting a tax deduction on its bank loan and also on the on-lending into the UK. For obscure technical reasons I think that wouldn’t work, but I’m aware some people take the contrary view… however I very much doubt anyone would go to that trouble for a measly £600k tax benefit.
There is a wider point here – ABP UK is (probably for historic reasons) an unlimited company, and unlimited companies aren’t required to file accounts, which limits our ability to see what it’s up to. That’s now an anachronism, and should change.
But – call me old-fashioned – I prefer not to accuse someone of “aggressive tax avoidance” when I don’t actually have the evidence.
-
1No, nobody in particular comes to mind
-
2insert tired joke about arts graduates working at the Guardian
-
3really I can’t stress enough how complicated these rules are, and how over-simplified a summary that is