Fixing social security will be complicated—there’s no escaping that

The answer to “neat, plausible and wrong” solutions is to combine them

September 08, 2016
Campaigners lay down a banner reading "What would you do if your income were taken care of?" in Berlin, Germany, 29th May 2016 ©Klaus-Dietmar Gabbert/DPA/PA Images
Campaigners lay down a banner reading "What would you do if your income were taken care of?" in Berlin, Germany, 29th May 2016 ©Klaus-Dietmar Gabbert/DPA/PA Images

Every human problem has a solution that is “neat, plausible and wrong” an American journalist said exactly 100 years ago. But sometimes when a problem grows knottier, it seems that the number of neat solutions multiplies. It certainly feels that way when it comes to the question of how to improve Britain’s creaking system of social security, a byzantine scheme of cash transfers which pays out over £200 billion each year.

As things stand, social security consists of dozens of different welfare benefits, variously available on the basis of present means, past contribution, compliance with conditions, family circumstances, age and residence. It operates on totally different principles for pensioners and non-pensioners. And it sits alongside a hidden system of “shadow welfare,” in the shape of tax reliefs, which like social security are intended to support household living standards. If there was ever anything in need of neat solutions, surely this is it?

Yet, the conclusion of a year-long study by the Fabian Society is that, while there are at least five neat and plausible solutions, they are all wrong—at least as single “silver bullet” answers. This is partly because of the complexity and heterogeneity of British economic and social life. But it is also because social security is designed to do lots of different things, which no single policy can achieve. In particular, cash transfers exist to redistribute money between groups of people—to equalise living standards and life chances, and prevent destitution and poverty. But they are also there to redistribute money over time, across different points in each of our lives to reflect variations in our lifetime earnings and living costs.

So what are the neat solutions, and why are they wrong? First, there is the fallacy of so-called “pre-distribution,” which contends that the need for social security spending can be greatly reduced with full employment, cheaper housing and a higher minimum wage. All of these things are, of course, desirable. But none of them are likely to lead to significant improvements in living standards, without more generous social security. We already have historically high employment, housing market interventions will at best slow the pace of rising rents, and the new “living wage” does surprisingly little to boost the household-level disposable incomes of poorer families.

Then there are four neat paths for reforming social security. There is a school that wants to see near-total privatisation, with almost everyone expected to provide for themselves with Singapore-style lifetime accounts and insurance policies. There is the Iain Duncan Smith approach, of creating a perfect system of means-testing while stripping away almost every other source of support. This is the philosophy behind universal credit, which dominates government policy for children and working-age adults. The third route for reform suggests we should revert to the social insurance principles designed by William Beveridge, which still work well for pensioners but have been abandoned for younger age-groups. Then, finally, there is the idea of sweeping everything else away and paying us all a flat-rate subsistence payment—the universal basic income.

Each of these ideas attracts ideological objections, but the Fabian research also reveals that none of them could really work in practice. Private welfare is inequitable and inefficient, but for most people it is also simply unaffordable. You only need to look at the post-2012 student loan system to see an example of a self-funded public provision that places totally unrealistic financial burdens on citizens. We also found that the cost of anything more than a few weeks of joblessness is beyond most people’s capacity to save or borrow.

Meanwhile insurance-type solutions offer a very partial answer to supporting children and working-age adults, whether they originate from the public or private sector. This is because most non-pension social security spending does not address risks, like short-term unemployment, that can be easily insured. Instead the lion’s share of the money goes to people who will be out of work for a long time, because they are disabled or care-givers, or to those with earnings which are insufficient to meet the costs of children and housing. This explains why Beveridge’s social insurance only succeeded for pensioners while the UK has become so reliant on means-testing for other age groups.

The government’s own neat solution has been to create universal credit, a single, simplified means-tested benefit, designed to defeat the “poverty trap” by creating clear financial incentives for every recipient to work or earn more. The rhetoric has been evangelical, but the reality is that the new benefit will be barely better than its predecessors at making work pay, while also being much less generous for most recipients. Nor does universal credit solve the underlying divisiveness of means-testing. Even though most of us will benefit from means-tested support during our lives, people still perceive it as a regime of “them” and “us.”

This is ironic, because the cash support which the government provides in tax relief, to people with sufficient earnings, is now pretty much as generous as our increasingly threadbare system of means-tested benefits. Taking the two together, we have a quasi-universal, flat-rate approach to supporting living standards, but one that sets people apart instead of binding them together. So why isn’t the answer to replace them both with a universal basic income? It is a long-discussed but newly-fashionable solution to the problems of means-testing, which would make work pay, reduce state intrusion into people’s lives and could be a long-term response to the possibility that work is less secure or plentiful in future decades.

"Universal basic income is a theoretically elegant policy—that might make matters worse"

But the Fabians and others have demonstrated that, for all these attractions, a fully-fledged UBI cannot be the answer. It would have to overcome huge barriers of public acceptability, in seeking to do away with the principles of contribution (“pay in more, get more out”) and condition (“seek work, if you can”). And it would not really eliminate means-testing, because you would still need a separate system to help people meet unaffordable housing costs.  

But our main objection is that it would do little to help poorer families: a full UBI is not an egalitarian policy. This is because a British UBI would be set at the same rate as universal credit, which by the 2020s will provide a totally inadequate income for many households. At the same time a full-time low earner would end up no better-off, because for every extra pound of basic income, he or she would lose a pound in tax relief (and that’s before thinking about broader tax rises, which might be needed to fund the scheme). Modelling by the pressure group Compass (which set out to support a UBI) showed that a full version of the scheme would actually drive up poverty. The priority must be to boost the disposable incomes of the millions of people in the UK struggling to make ends meet today, not to create a theoretically elegant policy that might make matters worse.

The Fabian report is not just a knocking exercise however. It is profoundly optimistic and shows there are many positive ways forward, as soon as we ditch the search for a single neat solution. The truth is that all of these ideas have something to commend them, but all fail on their own. Our answer is to design a blended system for the 2020s, which combines something of them all, while spending roughly the same share of GDP as we do today. This new system would contain tiers of support included a modest flat-rate credit in place of tax allowances, better contributory benefits during working life, limited private sector involvement, and more generous household-based means-testing too.

Taken together, this would create social security that is much more generous for poorer households and better too at distributing resources across everyone’s lives. No single policy can achieve both these ends, but together many can. The answer to “neat, plausible and wrong” solutions is to combine them.

For Us All: redesigning social security, for the 2020s by Andrew Harrop is published by the Fabian Society