Hervey Gibson: Secretary of State's fudge does little to help debate

Last Wednesday, Michael Moore laid before parliament a paper with 27 paragraphs and a table of numbers, saying it showed the effect of devolving corporation tax (CT) would be to "reduce overall spending power by around 10-12 billion over five years". It's a lulu.

Even at face value, the five-year total should be less than half the figure Moore quotes.

The spreadsheet makes the bizarre assumption that Scotland's new tax rate of 12.5 per cent was introduced four months ago, on 1 April, 2011. It is obvious why it does this, because it makes numbers that HM Revenue & Customs (HMRC) considers frightening look nearly 30 per cent bigger, but it's ridiculous. There could be no question of putting the changes through both parliaments and the HMRC bureaucracy before April 2013.

• Download a PDF critique of HMRC's estimates of the cost of

reducing Corporation Tax in Scotland

I have gone line by line through the calculations. I have sent my critique to Moore and it can be downloaded from The Scotsman website. The first and largest line over-estimates the cost to HMRC by nearly one-third, because it assumes taxes will come down from 26 per cent to 12.5 per cent, whereas the correct assumption is from 23 per cent to 12.5 per cent.

Line two assumes Chancellor George Osborne's tax reductions will apply only to large firms, not to small ones. Nothing has been announced, but I'd like to see him try that at the Tory party conference.

The paper estimates 18,000 businesses will be incorporated just because the CT rate comes down. From HMRC's point of view that is bad, because they will pay less tax as businesses than their proprietors would pay as self-employed.

To justify the 85 million "loss" HMRC postulates, these firms will have sales of 5bn. If this astounding forecast was a considered estimate, I would congratulate Moore on discovering the best and easiest business-creation mechanism known to economic development.

HMRC makes assumptions about "tax shift" - companies moving their headquarters to reduce tax bills. It tailors the assumptions to suit the context; low when it's against its case, high when it thinks it's in favour.

So it underplays the 250m benefit to the UK economy of businesses moving nameplates from abroad.

It overplays the amount nameplates would move from England to Scotland because it think that's a cost - but to Scotland it would be a benefit. The difference is 1.2bn a year. It also thinks Scotland's contribution to UK tax revenues today could be 250m more than the Scottish Government's GERS estimates. At one stage, Moore gives a subtotal, a "cost" to the Scottish budget of 1.8-2.1bn.Then come two lines of transparent fudge, adjusting for enhanced financial sector profits, which could only be right if Royal Bank of Scotland made full-year profits of about 4bn.

But the HMRC paper doesn't find any of the positives, of which the Treasury found rather a lot when it looked at Northern Ireland. The coalition consultation paper on CT for Northern Ireland nominates 13 points. This "report" addresses just one of those - and gets it wrong.

Not to be disingenuous, what seems to have happened here is that, stung by our remarks on spreadsheets and rhetoric, Moore has asked HMRC to come up with a proper paper. And, to save his face, please could the number in the bottom-right-hand-corner be more than 2.6bn?

The Treasury guys have done what was asked, but they have looked at the issue strictly from HMRC's point of view. Moore has proudly laid it before the House of Commons, without realising the interests of Scotland and HMRC may be diametrically opposed.

The Secretary of State and HMRC have been drawn into opposing devolution, even before the evidence has been prepared, presented and weighed. That is constitutionally unfortunate, politically misjudged, and is not going to help us have a useful and fair debate.

• Hervey Gibson is chairman of Cogent Strategies International and former head of economics at Scottish Enterprise.