GEORGE Osborne told the Commons the 50p top rate of income tax “raises at most a fraction of what we were told, and may raise nothing at all,” while causing “massive distortions,” justifying, he said, cutting the rate to 45p from next year.

Can it be true that high tax rates raise so little?

If high-earners did not change their behaviour at all when the tax was implemented, HMRC estimated 300,000 people would have paid a total of £6.5bn extra in 2010-11.

However, the vast majority of the relevant income was simply attributed to 2009-10, known as “forestalling,” paying the 40p rate and declaring less income under the 50p regime. That cut the yield from an expected £2.5bn to under £1bn – so at first glance, the tax does indeed appear to be a flop.

However, this forestalling is only possible in year one – if the tax was permanent it could not be repeated, so we would see how much was paid in a “standard” year and so how sensitive high-earners are to tax changes, generating an accurate Laffer Curve model on which to estimate the real impact.

The curve initially used by the government was very steep, suggesting a tax rise would generate a lot of cash.

As this did not happen, the curve used was amended (pictured left). On this shallower curve, the government hopes a small tax cut will barely impact revenues as it boosts incentives to work and pay the tax.

However, as we have seen this is based on a very unusual set of figures, with a huge deal of forestalling, and so may be as inaccurate as the initial curve.

The model Osborne has used is at high risk of being severely flawed.


We will never know exactly how much the 50p rate could have raised – its introduction and cut were both pre-announced, allowing savvy high-earners to shift earnings into different years, avoiding tax and skewing the figures.