Redistribution, social insurance and fairness

Published date
16 Jan 2018

Paul Johnson, director of the IFS, wishes politicians would stop talking about ‘fairness’. Writing for The Times last week, he argued that ‘it obscures many fundamental questions such as what rules should govern the market and how big a role we want the state to play in our lives. That’s what we should be talking about.’

One of those questions is the extent to which we see the role of the welfare state in terms of social insurance or redistribution. Johnson argues this what is really at stake in the controversy over how far people should be expected to meet some of the costs of long-term care out of their own assets.  

‘If you think the state is there to provide a degree of social insurance, stepping in where private insurance markets don’t work to pay for those who are unlucky enough to need care, then you are likely to favour it paying all the costs above a certain level. That’s how we tend to think of the NHS. But if you think the state is there just to redistribute money from rich to poor then you might think [capping the cost] unfair.’

Of course, Johnson is slightly simplifying here: few people are likely to think that the welfare state is there just to serve one of these purposes to the exclusion of the other, and most welfare states combine redistribution and social insurance to varying extents.  But the distinction he is making is fundamental to the purposes of welfare states and is terribly neglected in UK political debate. It goes far beyond the issue of long-term care.

Put crudely, redistribution is a zero-sum game in which, in order to make anyone better off, you need to make someone else worse off. (Ideological opponents of redistribution go further of course: they claim that redistribution leads to lower economic output, so it is in fact a negative-sum game where society as a whole is made worse off.)

Social insurance, provided it is well-designed, is a positive-sum game.

As long as people are concerned about risks which cannot be insured in private markets – and for most of us that is most of the big risks to our living standards that we will face over the lifecycle –  social insurance can make the great majority of people better off. (Not everyone, because there will usually be some people who will be wealthy enough to provide for themselves and who would therefore be better off opting out.)

This is probably how people tend to think of the NHS, as Johnson suggests, but the point applies more broadly – including to social security, which in UK political debate is routinely treated as a system of pure redistribution to the complete neglect of its insurance functions.  

Research by the IFS indicates that the insurance value of social security  over the lifecycle meets the positive-sum criterion comfortably: for the lowest two deciles of the lifetime income distribution, insurance via the tax and benefit system is worth more than five per cent of  lifetime income; for the middle of the distribution, about 3%; and gains from insurance are positive up to the 9th decile of the distribution (based on the 2014 Universal Credit rules compared to a hypothetical distributionally neutral baseline).

The positive-sum aspect of social security features little in contemporary public debate- to suggest that ‘welfare’ might make the majority better off sounds positively eccentric. Part of the problem is that the benefits to the majority can really only be measured over extended periods – ideally, over the entire adult lifecycle, as the IFS research aimed to do –  while at any point in time the system appears simply to be transferring income between groups in a zero-sum fashion, leaving plenty of scope for the ‘deservingness’ of claimants to be called into question.

At the same time a large part of the insurance value of benefits and of other forms of welfare state provision is non-cash in nature. Insurance reduces the losses people face in the event of certain contingencies arising. This protection is of value to them whether or not the contingencies actually occur.

This is well understood in the case of healthcare (we are generally happy to pay for NHS coverage but would prefer not to have to avail of it), but this understanding does not seem to translate to other welfare state functions.

There is also an institutional aspect: although the UK welfare state provides a lot of insurance, unlike many other countries – and unlike its own original Beveridgean blueprint – it does not generally do so as part of a formal social insurance system.

National Insurance contributions are not clearly distinct in purpose from general taxation, and redistribution and insurance are tangled together in the same benefits. Tax credits redistribute income towards lower-income families: but they also insure against falls in earnings, as many people discovered during the last recession. Untangling positive-sum social insurance and zero-sum redistribution is not a straightforward task.

This is not necessarily a problem in terms of how well the welfare state fulfils either function, but it does have an effect on public debate.

The slogan of ‘fairness’, which almost inevitably imposes a zero-sum framing on debate- certainly isn’t helping.

In a period when new and not-so-new forms of insecurity loom so large in the political agenda and in people’s lives, a greater focus on the insurance aspect of the welfare state, on its ability to pool risks as well as to transfer income, is long overdue.