Andrew Mobbs (mobbsy) wrote,
Andrew Mobbs

So, the spending review… the "Distributional Impact Analysis" that forms Appendix B of the Spending Review 2010 report makes interesting reading.

Just going by their own figures:
Table B.1: Weighted average annual net equivalised household income and benefits in kind by quintile
Q1 Q2 Q3 Q4 Q5
Weighted average income £13,800£19,100£24,200£31,700£48,700
Weighted average 2010-11 benefits in kind£11,500£10,700£7,800£6,700£5,400

Table B.2: Table of absolute weekly changes in household benefits in kind by net equivalised income quintile
Q1 Q2 Q3 Q4 Q5
Absolute weekly change in benefits in kind£7£10£11£10£10

They then go on to produce a chart (Chart B3) of percentage change in benefits by quintile, with the ridiculous claim:
In order to understand the relative impacts of the spending plans, it is useful to consider these changes as a proportion of the initial benefits in kind consumed by households. Chart B.3 illustrates this result. Quintile one has the lowest proportional change and quintile five has the highest.

I can't see how looking at proportional change in benefits is at all helpful for understanding the impact of cutting them when the requirement for benefits is so disproportionate across income.

It'd seem much more reasonable to me to chart the percentage change in income caused by these changes, not benefits. The definitions are a little obscure, but I'm fairly sure from the data sources annex, and comparing with NSO data that the given incomes include the effects of benefits.

So, just looking at yesterday's spending review benefits changes, the results on income look like this:
Q1 Q2 Q3 Q4 Q5
Weighted average income £13,800£19,100£24,200£31,700£48,700
Absolute weekly change in benefits in kind£7£10£11£10£10
Annual reduction in income (52x weekly)£364£520£572£520£520
Percentage drop in income2.6%2.7%2.4%1.6%1.1%

Really a rather regressive picture.

The mitigation is given in that the earlier announcements of increased taxation from the March and June budgets will, apparently, more than compensate for this regressive set of changes in benefits. Unfortunately, we can't check this claim since "These figures are calculated as economic estimates,
including the effects of assumptions and results from economic analyses that have a material impact, and are therefore outside the domain of official statistics."

However, even the Treasury's final consolidation of all changes, shown in Chart B6, which is admittedly rather heavily qualified, shows a regressive picture, and I think it's worth reproducing here:

It's worth noting how skewed the top quintile is by the extremely rich. I think aldabra was quite right to use Q4 as a proxy (even the majority of people in Q5 will be more similar to Q4 than the top end of Q5. For example, capping pension contribution tax relief at £50,000 doesn't actually affect people in the 9th decile at all, and probably only a small proportion of those in the 10th).

So, the people who lose the least out of all these changes are the moderately well-off, definitely 4th quintile, and I'd assert the same would be true for much of the 5th quintile and the upper end of the third.

While people don't (often?) collect information about income and polling intention (though if anybody has data, let me know), they do about "social class", which is determined by job. The results are unsurprising that as social class increases, so does both likelihood to vote and chance of voting Conservative (Lib Dem voting intention is fairly flat across class, except for a dip in C2 "skilled working class").
  • Post a new comment


    default userpic

    Your IP address will be recorded 

    When you submit the form an invisible reCAPTCHA check will be performed.
    You must follow the Privacy Policy and Google Terms of use.
  • 1 comment