IFS and the 50p tax rate – expressing massive doubt about the Treasury estimates

Posted on

The IFS have published their budget commentary.

I've looked at the 50p tax rate commentary. It says:

- No administrative data on how much actually paid until tax returns for 2010—11 submitted in January 2012

So the claims made are not based on actual data.

It goes on:

- In Budget 2011, Chancellor asked HMRC to produce a report on how much 50p rate was raising in time for this Budget

What's clear is we did not get it. Instead we got a figure based on 2009 analysis:

- HMT’s estimate from 2009 implied significant reduction in income resulting from the change: static costing £6.8 billion

So first, HMRC did not report on tax return data per se, they instead made a guess on what IFS thinks an inadequate base of what tax might have been paid if 40p rate had been still in operation - and they describe the data set used as 'the best they could do with information available' but clearly imply that does not mean it is reliable.

So all we actually got was an HMT guess, and nothing like facts.

The IFS note two possible responses:

Genuine reductions in income: working less hard, retiring early, leaving the UK

— Would expect these to reduce expenditure and hence indirect tax revenues as well as income tax revenues

In other words this footprint should appear in more than income tax. There's no sign it did.

Second possible response was avoidance. They say:

Avoidance responses: shifting income to different time periods, converting to capital gains, shifting income between spouses
— Would probably not reduce indirect tax revenues
— Moving income forward from 2010—11 to 2009—10 particularly important when examining 2010—11 data: clear incentive to do this when tax rise pre-announced

Seems that the latter is what was found. All they can say for sure is:

-  Overall, incomes among those with incomes above £150k increased 14% in 2009—10 but fell 25% in 2010—11
— Particularly for dividend income: grew 78% among this group in 2009—10 and then fell 73% in 2010—11
— Incomes would have been much lower in 2009—10 without 50p rate, and much higher in 2010—11

So there was massive avoidance but that does not mean it would be repeated.

 After that t's all down to modelling - the results of which the IFS says could be highly unreliable - a conclusion they apply to their own work as well. Certainly the data they present makes forestalling look like by far the biggest explanation for the loss of income and makes other factors look small - which was HMRC assumption in data I have used for my estimates and which I believe to be right - because in many cases the room for manoeuvre on income will be small due to contractual and other constraints.
As they say:

Unclear to what extent this was responsible for significant drop in income in 2010—11
— Using reasonable assumptions, HMRC find other considerable behavioural responses that reduced incomes
— But model is so imprecise that results are not very informative in themselves

In other words, alternative interpretations are viable. I suggest mine remains so. And I also suggest they remain within the plausible range for 2011/12, which is the year they were estimated for.

It will be curious to see what the data eventually says. I am very sure it will not be that the 50p rate raised anything as little as £100 million - or that incomes fell to suggest that was a significant behavioural response. But by then it will be too late.