To explain the tax gap to Richard J Murphy

The UK loses 20% of total corporate profits to tax havens but HMRC are in denial about the missing £7 billion

Sigh.

Gabriel Zucman has estimated the loss of corporate tax revenue due to artificial profit-shifting.

The UK is losing over 20% of all corporate profits that should be subject to tax before they ever get near the tax system if this data is right (and Zucman tends to understate things in my opinion).

The question is, what is the definition of “artificial”?

Zucman would say that Google selling advertising into the UK from Ireland is artificial profit shifting. The EU would say that’s an inevitable part and parcel of the Single Market rules, where one single company, in any one EU country, can sell into each and every other EU country while paying corporation tax in just the one – absent the existence of a permanent establishment.

HMRC also says that a corporation selling across borders, into a state in which the selling organisation does not have a permanent establishment (and yes, HMRC does indeed distinguish between the actual legal entity doing the selling and other parts of the same group which might have such an establishment – for that’s what the law actually says) is not tax avoidance. They state, as they should for that is what the law is, that this is just the way the law works.

That is the summary of the latest HMRC tax gap data. I have highlighted the large business line in blue: they’re the ones who, by definition will be shifting profits out of the country.

HMRC say the gap by these businesses is 5%. So, you could argue that using the Zucman data HMRC are 15% short. But I would argue that the 5% is based on the profits stated after the shifting Zucman has recorded has taken place. HMRC do not recognise profit shifting as tax avoidance: they have persistently said so. In that case extrapolating their data, if the 5% loss is stated out of 80% of the total (i.e. the sum net of Zucman’s 20%) then it’s 4% of the real total, meaning that the tax gap Zucman might be identifying is worth at least five times as much, or £7 billion a year on top of the £1.4 billion HMRC already recognise.

Some obvious thoughts follow. The first is why is HMRC in denial on this?

The second is that if one tax gap is wrong by a factor of five how much are the others out by?

Third, if they’re out by anything like the same amount why isn’t this creating more of an issue?

And why, in that case, isn’t HMRC’s failure to get something so basic right the cause for censure in its own right?

And why isn’t action to recover this sum resulting?

The difference, dear Senior Lecturer, is that you are using a different definition of the tax that is due. HMRC is using the law. You are using the definition of “how much tax Richard J Murphy thinks the bastards should be paying.” Amazing, I know, possibly even a little ego deflating, but the definition used in law is the one laid down in law, not something dreamed up by a retired accountant from Wandsworth.

Google Eire is entirely, absolutely and completely, obeying the law by selling advertising into the UK from Dublin. You want to claim that’s tax avoidance and no one else at all does. There’s the difference.

Gabriel Zucman also makes a large mistake. He’s not grasping that in order for those Google profits to be paid out to shareholders they must first enter the US corporate income tax net. And pay that tax. The entire structure delays a tax bill, to be sure, but that’s all it does. And that’s also how the corporate taxation system works – tax should be paid where economic value is created. Which in the case of an American company employing Americans in America to write a search engine is going to be, at least in part, in America. Under whatever American tax rules are.