The Daily Mail this morning reported that "Labour's 50p tax rate 'could cost the country £350bn', Osborne warned". The claim was calculated on the back of a fag packet.

The Daily Mail this morning published the surprising news that “Labour’s 50p tax rate ‘could cost the country £350bn’, Osborne warned“. The number, if it were true, would be eye watering. Readers can decide for themselves whether it’s worth more than the fag packet on which it was calculated.

The source of the story is the Adam Smith Institute who have put together a short report titled ‘The Revenue and Growth Effects of Britain’s High Personal Taxes‘. The report concludes:

“At a minimum, the policy risks flat growth for a decade, and a possible recession at the end of the period. The net effect upon government finances, adverse from the outset, gets worse from year to year and ends up with net losses of some £350bn after ten years. In the worst-­‐case scenario, approximately £640bn of revenue would be forgone.”

The Mail made no mention of the fact that the £350bn would be spread over 10 years but, even so, £35bn is a huge number. The Adam Smith Institute get this number and their claim about “flat growth” from projections over how many individuals will leave Britain as a result of the tax regime. As the table below shows, the survey of “tax advisers” found that at a minimum, 25% of personal taxpayers and 15% of corporate taxpayers are considering leaving the UK.

From their survey of “several dozen” tax advisers (I’m told by author, Miles Saltiel) and assumptions about migration and recruitment, which are not disclosed in the report, the authors extrapolate that at least 17% and up to 31% of higher rate taxpayers working in London’s financial services sector will leave the country. In other words up to a third of all those working in the City and earning above £150,00 will have left the UK by 2016.

Saltiel told me that his calculations were “a lot better than a wet finger in the air” and that although he was “pretty comfortable” with the conclusions, there were not enough survey respondants “to do a statistical thing”. Statistical things aside, readers will no doubt remember similar claims by Paul Daniels and Frank Bruno after the 1997 election.