The second half of the IFS briefing was all about the distributional effect of the Spending Review. And you know what that means: decile charts – and lots and lots of them. As it happens, there were some areas of agreement between the IFS and the Treasury figures. Both, for instance, say that the welfare measures set out in yesterday's Spending Review will affect the least well-off the most. But there was one main area of disagreement. The Treasury says that its combined tax and welfare measures up to 2012-13 will be broadly progressive. The IFS says that they will be regressive.
This is exactly the same issue that cropped up in August, when the IFS was commissioned to write a report by the End Child Poverty Campaign. And it comes about because they model certain tax and benefit changes differently from the Treasury, to produce the graph below. It's the white line (measuring the impact of tax and benefit chages on percentage income) that's important – the IFS describe this as "slightly regressive or flat within the bottom nine-tenths of households."
When the horizon is expanded to 2015, the IFS trend becomes more marked. Here they describe the white line as "clearly regressive within the bottom 90 percent," adding that "the regressive impact is the result of reforms announced by the current Government both in the June Budget and in the Spending Review". Here's the chart:
But these points were made far less aggressively than they were back in August. As one audience member put it, there were a lot of "ifs and buts" on the IFS's part. They emphasised, rightly, that tax and benefit changes alone don't cover the full scope of the government's budget measures. And they congratulated Osborne for attempting to include the effect of public service cuts in his charts, even if only a third of the cuts can be modelled in this way. This less strident approach was encapsulated by the IFS's Carl Emmerson in his opening address: "This is not to say that [the Spending Review] is unfair: fairness will always be in the eye of the beholder." Quite.