After months of nudges and winks, the UK Coalition has finally started cutting middle-class benefits with the totemic withdrawal of Child Benefit from higher-rate taxpayers. Here I want to look into what effect the evidence tells us this will have on inequality – and suggest some crucial issues on which the evidence doesn’t really tell us anything at all…
The current consensus…
There are three things we know about how means-testing compares to universal benefits. Firstly, if you’re a conventional economist, you’ll tend prefer means-testing – payments are meant to tackle poverty, and universal benefits ‘inefficiently’ give money to rich people.1 This is what the Work and Pensions Secretary Iain Duncan-Smith means when he says it’s ‘bonkers’ that people earning £50,000 receive benefits.
Before concluding that means-testing is the answer, we need to consider both ‘practicalities’ and ‘politics’. Practically, some people don’t claim means-tested benefits (due to a combination of complexity, intrusion and stigma), and those that do claim them will see that they are penalised for earning more. For Child Benefits, the level of take-up might be less affected – the means-testing is based on the already-existing tax code – but incentives to earn more will deteriorate.
What I want to concentrate on for the rest of this article, though, is the problem of politics, and particularly the classic ‘paradox of redistribution’ noted by Walter Korpi and Joakim Palme (journal version here). Put simply, the more you target benefits on the poor, the less effective you are in the long run in reducing poverty and inequality. This is shown in the following graph (as long as you bear in mind that lower values of their ‘index of targeting’ mean that transfers are targeted at people on low incomes…):
Why does this happen? Put simply, it’s because countries with less targeting have bigger welfare budgets overall. Targeting welfare affects how we talk about both the payments and the people that claim them. For example, Child Benefit was originally set up to make up for the abolition of the child tax allowance – itself a way of showing that society recognised that bringing up kids was expensive and socially desirable. The rhetoric today was much more about poverty reduction – and in the long run, scare stories about people who don’t ‘deserve’ the benefit will cut the generosity of the benefit. (You can read a wonderful, extensive discussion of this in the Fabian’s ‘The Solidarity Society’).
What not to cut
When we come to apply this to Britain today, though, we’re faced with a problem: what do we do when we have to cut government spending by an unprecedented amount?2 For those who want lower inequality, the usual answer is that it’s better to keep the universal welfare state and instead raise taxes – but given the scale of the current cuts, no-one is seriously suggesting that tax rises can solve the problem on their own.
What we really want to know is: which bits of the welfare system are most important to keep universal? Is it better to keep Child Benefit universal, or the Winter Fuel Allowance given to every pensioner? Indeed, is it more important to keep welfare payments universal, or to avoid means-testing/targeting in for hospital treatment, or dentists, or universities, or anything else?
Somehow there is little evidence to guide us, but I’ve managed to pick up a few ideas over the past year…
- The idea of ‘progressive universalism’ has been increasingly popular in the UK – you give everybody something, but more to the worst-off. This could have been achieved by taxing Child Benefit for everybody, instead of completely removing it from higher-rate taxpayers. Whether this maintains solidarity to the same extent is unclear, but is worthy of further study.
- Some universal services have wider effects on social solidarity. Mario Luis Small has shown how women often form links with other women at children’s centres in the US, with positive effects for both them and their kids. The Surestart centres in the UK have not only improved outcomes for children, but are likely to have helped ‘community’ too. I think restricting Surestart to poor areas would be particularly damaging to social solidarity.
- Stigma and low take-up are likely to be lesser problems for socially desirable services like higher education – where Timo has argued on this blog that there are other fairness reasons for introducing means-tested charges.3
These partially-formed ideas suggest that there is something important to be said about targeting universalism – but that it hasn’t been said yet. Hopefully someone will fill this gap before the present crisis erodes universalism across the British welfare state beyond repair.
3. I can’t claim any credit for this thought – thanks to Irv Garfinkel for suggesting it to me.