There's much talk about Barclay's being able to reduce its tax bill if it moves its head office to the Netherlands as a result of any merger with ABN Ambro. This is unsurprising. The Netherlands is an aggressive tax haven, as I and my co-authors at SOMO have shown in our report, published late last year.
But there's another dimension to Barclay's tax that is as interesting. First of all, journalists are taking the tax figures on its profit and loss account at face value and saying Barclays pays tax at 27.2% as a result. This is wrong. It isn't. That includes deferred tax. By and large, as I've shown, deferred tax is not paid even when it is provided so that's a red herring.
But the much more important question at Barclays is a simple one. It's this. How come it doesn't pay its tax?
2005 is the last year for which we have audited results. At the start of that year it owed £621 million in tax per note 24 of its accounts. It had no tax assets. At the end of the year it owed £747 million, again with no assets.
In the year it had a current tax bill owing of £1,524 million according to note 10. There is no record there that any was due by an associate or any other party, although note 26 suggests £6 million at most might arise for that reason. That's immaterial in considering the numbers noted below.
According to its published cash flow Barclays paid £1,082 million in tax in the year.
Now there's a simple rule in accounting. If you take the opening tax bill, add on the current year's tax charge and take off the amount paid you should come to the closing tax liability. So in this case £621 million plus £1,524 million less £1,082 million should come to £747 million. Except of course it doesn't. It comes to £1,063 million. That's £316 million more than Barclays says it owed at the year end.
Now, that means one of these four numbers is wrong. So I hunted and I found that the tax bill as per the profit and loss account is wrong - it should be £50 million lower as that much tax receipt is recognised in the Statement of Recognised Gains and Losses. I've decided to ignore £27 million cost in the revaluation reserve (note 40) though because I think it's deferred, although the note does not make it clear. That does decrease the unexplained gap to £266 million. Which is better, I guess. But is still a mighty big number to 'lose' somewhere in a set of accounts.
And I stress, that's what happening here. As a reasonably informed chartered accountant I've used all the data Barclays have given me and it seems to make no sense.
If Barclays is right and it owed £1,524 in the year that's an effective rate of 28.9%. Which looks reasonable. On the other hand, and assuming balance sheets are right (as they usually are) and cash flows should be correct (because a bank should be able to tot up its books) then to make Barclay's double entry work it could only have owed £1,258 million on current account (£50 million being credited off the P & L account). That's an effective rate of 23.8%. The £266 million difference has been simply 'lost' in Barclay's tax accounting.
So the real question is, how much tax does Barclays owe, and how does it account for it? Neither is obvious from the published numbers.
Thanks for reading this post.
You can share this post on social media of your choice by clicking these icons:
You can subscribe to this blog's daily email here.
And if you would like to support this blog you can, here:
Well, it’s an interesting thing tax accounting you know Richard. And I speak from personal experience, having once been personally responsible for a £100m error in a set of published accounts of a listed UK group. I was doing the group tax accounting for their accounts and totted up group potential deferred tax and inserted a sub-total, got called away on a phone call and returned only to re-tot up the numbers, not noticing I was double counting the sub-total. The number spiked in 3 years from £50m to £100m and back down to £50m and was only spotted when my successor came to do the accounts next year. Now that admittedly was potential un-provided deferred tax, was not part of the double-entry so that the mistake did not pop up somewhere else and had no effect on the P+L or Balance Sheet. It did amuse me though that not a single City analyst noticed, or at least commented, on this at all and this was a very active group that was very heavily analysed at the time.
However, I’m guessing that this is not the sort of thing that happened with Barclays as clearly provisions and tax payments are part of the double entry. But then again, don’t Barclays have Roger Jenkins in-house who is worth £40m or so salary a year to engineer precisely this sort of thing for the clients of Barclays as well as Barclays itself?
[…] So they should. I highlighted the problem in ‘Do they add up‘ which showed conclusively that the tax accounts of FTSE companies do not do so. As Barclays have gone out of their way to prove. […]