Friday, 23 October 2009

Innumerate twat of the week

I trust we've finished discussing yesterday's QT (I am on the road so off line out of office hours) so allow me to nominate James Haldenby for this week's ITotW Award. From yesterday's Times:

At Reform we have calculated that the true amount of benefits claimed by middle-class people is £31 billion. That is £1 in every £4 spent on benefits and enough to pay for the police service twice over. The benefit most skewed to higher earners is statutory maternity pay, where £8 in every £10 is given to people on or above middle income. Child benefit, tax credits and the basic state pension are the other main offenders, with subsidised loans for higher-education students also benefiting the better off...

Our research sets the line at an income of £40,000 for a couple with two children (ie, £15,000 of income per adult plus £5,000 per child). Others would define middle class at a lower level of income, meaning that middle-class benefits cost even more...

Middle-class people tend to oppose higher taxes, but that is the price of our entitlements. £31 billion is the equivalent of an extra 8p on the basic rate of income tax. We are being bribed with our own money.


OK, let's assume that's all mathematically correct, where is the logical error? What's the missing figure? Why does no politician just do this? If it's such a brilliant idea, how come it's not already in the MW manifesto?

The answer is simple, he ignores the means-testing. Assuming that £31 billion is shared by the richest quarter of households, that's about £5,000 benefits each per annum. If they means-tested all these benefits so that a household with gross income of £40,000 received nothing, the benefit withdrawal rate would be about 12.5% (i.e. £5,000 divided by £40,000).

So households on incomes up to £40,000 might save 8p income tax on every £1 they earn but would lose 12.5p in benefits withdrawal for every £1 they earn instead, so they'd be worse off in net terms and have an effective marginal tax rate that is 4.5% higher (as well as getting involved in the parallel universe of means-testing), but people on very high incomes would be better off as they have a marginal tax rate that is 8% lower.

That may be the outcome that James Haldenby wishes (in which case he is a twat) or it may be that he hasn't thought it through properly (in which case he is still a twat).

7 comments:

James Higham said...

There was a nice one at Vox's about a CEO who can't add up too. Seems to be catching.

marksany said...

I'm surprised he doesn't want to take personal allowances off the "middle classes" too. After all giving you a benefit or giving you some of the tax you paid back are pretty
much the same thing. The elephant in the room is the effect onarginsl withdrawal rates and the massive disincentive this is to earn more.

bayard said...

Correct me if I'm wrong, but isn't the personal allowance simply there to stop HMRC having to send out tax demands for ridiculously small amounts?

marksany said...

"onarginsl" WTF did I mean by that? This iPhone smellchekker is a bit radon.

Mark Wadsworth said...

JH, link?

Marksany, indeedy. They already 'means-test' the age-related personal allowance and are going to do it for people who earn more than £100,000. I assume that the mystery word is "marginal".

Bayard, nope. The personal allowance is currently £6,475 so if you earn £6,480 you still get a ridiculously small tax demand.

bayard said...

Mark, ISTR that a small amount of tax due is simply knocked off your personal allowance for the following year, but yes, the personal allowance could still be a lot smaller and serve the same purpose.

Mark Wadsworth said...

Bayard, sure, but that is not the original reason why we have a personal allowance (you can code stuff out using K codes which mean negative personal allowances).

It's historic really - when income tax was introduced, it only applied to incomes over £100,000 (or today's equivalent).