Would UBI be the best use of the money it would cost?

I’ve been having a few conversations about Universal Basic Income (UBI) with philosophers. Its popularity among that crowd continues to surprise me. Its popularity among some crowds is often ascribed to its simplicity: it has the “one neat trick” appeal. But this sort of appeal works on people who like things simple and want to get through the thinking stage and into the acting stage as fast as possible. Academic philosophers are as far from that as it’s possible to be. They never want anything to be simple. They never want to act on a problem until they’re 100% sure they can’t make the issue any more complicated. If they had their way, there would be no acting stage at all, just an endless analysis of distinctions into further sub-distinctions, to infinity. I sit in committee meetings with them.

Part of the philosophical appeal of UBI might be that it makes a policy problem look like a philosophical problem. My colleague and collaborator Benjamin Sachs-Cobbe has made this point. In economic terms, there is no feasible UBI scheme whose outcomes couldn’t be achieved by the right system of conditional welfare benefits. UBI just sounds different, philosophically, because of the way it’s framed.

The reason for this is that UBI, in practice, can only mean Universal Basic Gross Income. UBI advocates sometimes trade on the “free money for all!” slogan, but of course this could only be in gross terms. In net terms, some people will have to pay more into the UBI scheme than they get out of it. There is no such thing as Universal Basic Net Income. If the UK government started paying every British adult £10k a year, that would cost £530bn. Even if the UBI replaced the entire forecast social security spending of £316bn, there would still be £214bn to fund.1 That means a lot of people will have to pay more in tax than they’re getting from their new UBI.

How would the government then decide who is to be a net payer into the UBI scheme and who is to be a net receiver from it? Simple: progressive taxation, based on existing income other than the UBI. Those with low incomes will be net receivers from UBI; those with high incomes will be net payers. So in net terms, this is not at all universal. It’s just a conditional benefit, which you get if you’re poor enough, don’t if you’re not, and pay for if you’re rich enough.

Notice how easy it is to replicate this with a means-tested welfare system. The condition for receiving welfare is simple: your income has to be low enough. There is no extra means-testing required for this versus the UBI scheme; both require the government knowing people’s incomes, to determine whether they are to be net receivers or net payers. The welfare system pays precisely the same to everyone as they would receive from the UBI in net terms. It therefore costs exactly the same and can be funded by precisely the same taxation scheme. The economic outcomes are, when all the accounting is reckoned, identical.

So what’s the difference between these? Only a philosophical one. One is a radical rethinking of the relationships between work and entitlement. The other is a mundane income-support welfare policy. Their effects are identical. A colleague told me that there’s a moral difference between paying money that goes to somebody else, and paying money that goes to everyone, even if you pay more than you get and somebody else gets more than they pay. Comments like this just confirm to me that I don’t understand moral philosophy. Or, possibly, I don’t understand morality.

But I think this helps to explain the philosophical interest in UBI. It is presented as a radical paradigm shift—a major philosophical disruption. But this is just an artefact of the bait-and-switch that powers the UBI marketing machine. “Give everyone free money” sounds radical, impossible, mind-blowing. Wrapping your head around it seems to require rethinking your most fundamental beliefs. Call the philosopher! But when you get past the gross/net confusion it turns out to mean: “Pay benefits to those with incomes below a certain level, tapering off at higher income brackets, and funded by progressive taxation”. Framed that way, it’s just a welfare policy tweak—more within the expertise of a treasury accountant than a philosopher.

That’s not to say that philosophers can’t still be interested, once UBI is reframed as an ordinary tax and welfare policy. But I don’t know many who are. The illusion of a grand conceptual revolution is too distracting. People are always trying to make the question as magnificently abstract as possible. Some like to talk about UBI in the context of an imagined future, in which robots are doing the majority of the work, and most humans are involuntarily unemployed. Without knowing any details, how are we meant to meaningfully discuss which tax and welfare policies would be desirable in this science fiction world? If we want to debate UBI, it is more fruitful to debate a concrete case, where we can actually know the details, where the devil resides. So let’s try this: should we want to implement UBI in my country of residence, the UK, right now?

The first thing we can ask is: how big a benefit are we talking about? People like to say things like “an income level sufficient to meet basic needs”, but whose basic needs? Currently I think it’s possible for somebody in the UK to be eligible for around £26k worth of benefits. Suppose, for example, that they have a serious disability and require housing support in an expensive location. They could get something like £1400/month (£16,800/year) from Universal Credit, if they had both the limited capability for work and the housing element. They could also get a Personal Independence Payment of around £187.45 per week (£9747.40/year), if they were on the Enhanced Rate and had both the Daily Living and Mobility components. That comes to £26,547.40. I’m not saying that this is at all common, but there must be at least some people on this level of benefits. And, given how expensive life with a disability can be, at least some of them must need every penny of that. So a UBI that covered everyone’s basic needs would have to pay at least this amount. Providing it to every British adult would cost £1.4trn—more than the whole Treasury budget.

More realistic would be something like the £10k mentioned above. That would be enough to cover most of some people’s basic needs. People with special needs would have to qualify for other benefits. I estimated that the added cost of a UBI at this level would be on the order of £214bn, but that was assuming that all social security spending was replaced. In fact, paying £10k per adult would require much of that spending to be retained, if we wanted to keep child-related support, disability top-ups, housing payments, etc. So it seems fairer to imagine that the added cost would be something more like £400bn. That’s not taking into account any lost output if people choose to work less now that they get UBI.

Could HM Treasury raise anything close to another £400bn of revenue? Possibly. Some rich countries tax over 50% of GDP; the UK taxes only 40%. HM Treasury could start by eliminating the Personal Allowance up to £12,570 and instead tax that income at 20% (low-earners would still be better off, since they’d be getting the UBI—if you currently earn £10,000 a year, you’d now earn £16,000). It could push up all other marginal rates by 10% (again, lower-earners would still be better off—e.g., if you currently earn £20,000 pre-tax, you’d be better off by £3743). These changes to income tax could bring in £150bn or so, assuming that too many high-earners didn’t leave the country.2 Maybe Treasury could raise the remaining £250bn from reformed capital gains, corporation, and land-value taxes plus VAT.

Supposing it did manage this, however, we’d have to ask: would UBI be the best thing to spend this new £400bn on? Imagine, instead, spending £200bn to double the NHS budget. Imagine spending £100bn to build around 500,000 new homes. That would still leave £50bn to nearly double the core schools budget, £20-30bn to spend on providing universal free childcare and giving longer parental leave, £20bn on meeting Net Zero targets, £15bn on taking a “Housing First” approach to homelessness, with some still left over for benefit-increases, tax-relief, debt-reduction, or whatever (there’d also be money for a job guarantee, which on some estimates might be revenue-neutral or even revenue-positive).

These would be transformative changes. If they worked, they would go a long way towards solving the UK’s most serious sociopolitical problems. If HM Treasury had the money to pay a £10k UBI, it would also have the money to do things like this. When could we imagine UBI being the best use of what it would cost?

Thinking about questions like this is, in my view, a better use of political philosophers’ time than pondering vague speculations about a robotic future. If philosophers want to be useful (and I don’t know that they should) then they probably need to cultivate their taste for boring details and wean themselves off grand concepts.