The policy changes of the British government’s recent spending review are set out in a Treasury report. Attached as an appendix is a ‘distributional analysis’. You can get the whole thing here.
The Treasury has also published an equality impact assessment. It has to be said it looks like a pretty poor effort. Since 2006, it’s been a legal requirement for the government to assess its own measures in this way. As it happens, the coalition didn’t bother to carry out any sort of equality impact assessment for the post-election budget; as a consequence – and thanks to the efforts of Yvette Cooper and the Fawcett Society – we may now see a judicial review of the budget (although not of the spending review). The Fawcett Society’s assessment is that the government’s policies will directly affect women much more than they will directly affect men. They may well be right. This Guardian story has more details.
By contrast, there’s no law requiring the government to make a distributional analysis of what it does, so at first glance it looks as though they’re simply doing us a favour with that one. Perhaps the coalition saw it as having potential to persuade us that the cuts will be fair. If so, it hasn’t worked very well: you don’t need to do any sums to see that the Treasury analysis omits to include the impact of cuts to housing benefit, disability benefit(s), tax credits and council tax benefit. The IFS has been doing some of the hard work with numbers – taking account of all the things just mentioned that should have been in there to begin with – and you can see their preliminary findings here. In short, the government says that their spending review is progressive, but they’re wrong: it’s regressive.
My own small beer assessment, echoing Alex’s recent comment over at Crooked Timber, is this: the distributional analysis codifies long-standing conservative prejudices in a novel way. In what follows I’ll focus on paragraphs B.6 to B.28 of the report, which is a section that aims to show the distributional impact of changes to departmental spending. Paragraphs B.16 and B.17 say:
The modelling shows that all households benefit substantially from expenditure on public services. The mapping of the baseline consumption of public services (benefits in kind) … demonstrates that the consumption of services is skewed towards lower income households. This is explained by the following:
• demographic factors: lower income groups contain a higher proportion of children and pensioners, who are the most intensive users of welfare services;
• other factors affecting need, such as long standing illness, which is reported more in lower income groups;
• the targeting and means testing of certain services, such as social housing, social care and free school meals; and
• differential use of private alternatives, including private schools and health care.
The first thing to note is the use of the word ‘benefit’ and the phrase ‘benefits in kind’. They’re being offered up as substitutes for the more familiar ‘services’ or ‘public services’. Normally, in the context of welfare, ‘benefit’ means money. So what’s the coalition up to here? A fairly frequent theme in my parents’ house (it came up about once a term) was that people who paid to send their kids to private schools were doing everyone else a favour. Twice over, in fact. Once by continuing to subsidise public schooling through taxation, and again through lessening demand. Call this the ‘what dad said when the school fees came due’ principle. The coalition has taken the first half of this prejudice closer to bureaucratic reification by calling a universal non-monetary provision a ‘benefit’ and assigning it a monetary quantity. It used to be that the rhetoric of privatisation was about choice. This was when services that few would dream of attempting to do without (electricity, say) were being privatised. We’re probably beyond that now. In the Tory future, it seems, privatisation will be about private services – 16-18 schooling, say – that many people will have to do without, simply because they won’t be able to afford them. Those public services that remain will be tagged with a number prefixed with a pound sign, to emphasise just how generous, just how god damned progressive the government really is.
I disagree with this attitude, of course, just as I disagree with the notion that poorer people are better off than you might have thought because they get to consume more services. This is the second half of the ‘what dad said …’ principle. That it’s now official can be seen in the paragraph I’ve quoted, and also in that it’s supposed – at least I think so – to form the basis of our understanding of the graph in figure B.3 (p. 95) of the report. The graph shows a slope across quintiles that’s progressive rather than regressive. There are two problems here. The first problem is this. The government’s argument is that since the reduction in the total of non-monetary provisions received by poorer households will be smaller – as a proportion of what they received before – than the reduction in the total received by richer households, the spending review is progressive. However, when you consider any single non-monetary provision in itself, it may be far from obvious that it has redistribution as its aim. For example, health care. It seems that we don’t give people medical treatment because they’re poor, by way of compensating them for their lot in life; we treat them because they’re sick. This certainly seems true at the point of use – NHS doctors don’t hand out income questionnaires before admitting patients – and I suspect it’s also true at just about every level of NHS administration, up to and including ministerial level. If this is what we all believe, then we shouldn’t count such a provision into a quantified calculation that aims to show an overall progression or regression. We’ve already decided that it isn’t to be measured like that.
The second problem also has to do with incommensurability. Even if you believe that you can include a non-monetary provision into a distributional analysis, you still need to decide what the numbers are supposed to attach to. The report’s authors seem to have used numbers that reflect costs. But what have costs got to do with distributive justice? What we’re evaluating is who’ll get to have a better or worse time of it than they did previously, which is something that needn’t be connected to provider costs at all. Something along these lines clearly gets at the consciences of the report’s authors, since they say:
… there are a number of caveats to [our] methodology. It does not reflect the value people place on the services they consume nor how effective those services are at delivering desired outcomes (for example, low value programmes that are being stopped are measured at the same rate as high value programmes that are being kept)
You can do your own thought experiment on this one. Say you get sick, such that you need complicated and effortful hospital treatment. You get the treatment and it makes you healthy again. A lot of money has been spent on you, possibly. Do you feel grateful? Maybe so. But do you feel better off? Do you feel richer? If so, by how much? And would you rather have had the money? There’s an entanglement with counterfactuals here. It makes itself felt in the apparently absurd construction: ‘low value … measured at the same rate as … high value’.
Finally, it’s worth noting that the report reheats the old Tory favourite: aspiration. The idea is to get us to think of the lowest quintile / decile households as being rich in potential, and as such, not really in poverty. Am I excessively sarcastic? Well, Nick Clegg calls it ‘long-term fairness’, and I don’t see how I could beat that for taking the piss. Others, including the IFS, have commented already. I’m confident this’ll be coming up regularly over the next couple of years, since the coalition clearly thinks it’s a powerful concept that they’ve got hold of, and the criticisms just don’t wash. After all, just look how successful they are, being exemplars of the ‘jam tomorrow’ ethos.