50% Again/the IFS
I have have a lot of time for the IFS. But that does not mean I have to accept their numbers without question. In particular Newmania, in the comments to my last post has been using the argument that they reckon that a 45% super tax would cost the exchequer money.
The IFS report is here.
They use a research report (dubbed BSS) in all of their estimates. Rather than simply take the headlines as repeated int he press it’s worth looking at the report in some deatil. Especially these bits, which didn’t make the headlines:
However, BSS point out that there is a great deal of uncertainty around this number. First, the paucity of data meant that the authors had to use a relatively simple approach, leading them to describe this estimate as ‘very preliminary’ and ‘tentative’ (BSS, pp. 15 and 18). The elasticity is estimated using changes to top incomes that happened during the 1980s, a period when the top rate of income tax was falling and when income inequality was increasing. This approach may then confuse responses to the policy with any underlying factor increasing income inequality, and this might mean that the estimated elasticity is too high. Furthermore, even if their approach to estimating the parameter was correct, the estimate was based on a small amount of data and is therefore subject to sampling error.
They also have a chart, below. This estimates three different (all possible) estimates of ‘taxable income elasticity’.
Note how on one of their estimates a 50% tax rate would raise over £1bn.
When examining any economic theory it is worth looking at the assumptions, the variables and the possible problems.