Deal or no deal?
Posted by Christie Malry on December 20, 2011 at 9:45 pm
Oh my days. A lot of complete crud has been written in the last 24 hours about how appalling it is that senior staff at HMRC would even contemplate cutting deals with the companies it is supposed to be enforcing the existing tax law against. Especially in austere times, HMRC should be doggedly pursuing every legal challenge to ensure that it maximises the country's tax take.
This is, I'm afraid, a load of naive hogwash. There's a simple formula to determine whether it's worth proceeding with a tax case:
If L x s - (1-L) x S > C then it's worth proceeding with the case.
Where L is the likelihood of the case succeeding, s is the amount at stake, S is the amount of similar cases in other companies that will succeed or fall with this test case and C is the cost of bringing the case to court.
Of course, all of L, s, S and C are estimates. It requires enormous amounts of judgement to estimate suitable values for these variables.
The big risk is that, in bringing and losing the case, you open up the watershed to lots of other companies who had previously accepted HMRC's tax treatment but now find themselves in for a free tax refund courtesy of the brave company that brought the test case. For example, in the case of Vodafone, it rather suits HMRC to let Vodafone 'get away with it', even for a supposedly sizeable sum because, if HMRC loses, the entire edifice of CFC taxation might come crashing down, and with it many billions of pounds of tax revenues might need to be refunded. Given that the word on the street is that the UK's CFC legislation is on fairly shaky legal grounds, it's hardly a risk worth taking.
It's worth remembering the formula when you read the gibbering analyses of left-wing commentators. Because their hatred of big companies makes them oblivious to a very important fact about business: that the future is uncertain. Dave Hartnett may well have been doing dodgy deals. But anyone in his position would have needed to have taken a view about the likelihood of success, the amount at stake in the specific case, the amounts at stake in wider cases and the cost of legal action. Therefore, necessarily, there would always be a point at which it would be better to cut a deal than to stand and fight. The mere fact that the Goldmans and Vodafone deals were done cannot be used as ex ante evidence that Hartnett was bent.
[...] December 21st, 2011 · No Comments L x s – (1-L) x S > C [...]
With great respect, S is surely the "number" of similar cases. It cannot be the "amount" of similar cases.
Actually to make the equation valid S should not be the total amount at stake in other companies but the amount that HMRC expects to collect if the case is not lost in court.
Of course L and C both increase when you can employ specialist tax accountants either internally or externally don't they? Which of course is an option not readily available to smaller companies or individuals with lower levels of income. The consequence of all this is that rich companies and individuals have lower effective tax rates, specialist tax accountants have higher incomes and the rest of us have to pay higjher effective tax rates to compensate for the loss of tax revenue.
Specialist tax accountants are bad for society? Discuss.
[...] profits via Luxembourg. HMRC settled with Vodafone due to fears that, if the case were lost, other taxpayers might be able to reclaim tax already paid. The risk seeming too great, HMRC decided to take a little rather than possibly lose a [...]