Tomorrow night's Panorama is essential viewing for readers of this blog. Made by John Sweeney it looks at just what happens in tax havens.
The answer at £18.5 billion is, I suspect, an underestimate, but it's always worth being cautious when it comes to these matters. Even so the loss is staggering: that's £18.5 billion a year lost to the UK Exchequer as a result of the abuse of tax havens by corporations and individuals resident in the UK.
Part is avoidance. A significant amount is evasion.
And of course I'm not na?Øve enough to think that this means all can be recovered. But I'm equally adamant that significant amounts could be.
Much of the tax haven abuse by individuals does, I believe, take place in our Crown Dependencies. This could be eliminated either by our major banks being willing to comply with money laundering rules in those places, meaning that they would report all those who refuse to information exchange with the UK as potential money launderers. That would cripple the evasion activity in those places.
Or those places could join the full EU Savings Tax Directive and exchange information with other EU members. That would have the same effect.
And the UK supporting and imposing the expanded EU STD on them would be even more effective, but I'll tell you: my information suggests the UK is doing all it can to prevent that right now.
And of course, we could abolish the domicile rule which is, in any event, illegal because it discriminates on the basis of national origin contrary to the requirements of the Race Relations Act as amended in 2003. That would raise up to £4 billion.
And we could put our own house in order and create a Financial Transparency Act requiring the full accounts of all UK companies and trusts to be put on public record, with full details of beneficial ownership and management attached. It's not unreasonable to think that small company tax yield could increase by 10% of the corporate tax take as a result - so abused is the UK company register at present.
The list could go on. The important point is this. No one has dared calculate such a number before. I have done so. It is, of course, an estimate. It seeks to measure what is not recorded. That means it is bound to be an estimate. I happen to think it an underestimate. But it gives us an understanding of our loss and a target for our gain.
Isn't that vital if action is to be taken?
Now Alistair, what are you going to do about it?