Guaranteed minimum income: how much would it cost? (updated)

We were talking not long ago about universal basic income policy, and there were a variety of opinions about the desirability, political sustainability and implications of such a policy. But, before arguing about those issues, it’s useful to consider whether a basic income is feasible at all and, if so, what kinds of tax policies, and adjustments to other welfare policies, would be required to support it. I’ve considered the relatively easy case of a guaranteed minimum income, rather than a universal basic income paid to everyone, as advocated by Philippe von Parijs and others.

I’m going to work through it with a more or less realistic numerical example, with numbers chosen to simplify the calculations. The first simplification will be to consider only adults (I’ll try to explain my reasoning for children if I get time). I’ll assume that, initially 60 per cent of the population is employed and also that 60 per cent of income (before taxes and transfers) goes to labor, while 40 per cent goes to owners of capital and land. Initially, taxes are 30 per cent of total income, borne proportionally by labor and capital. For the moment, I’m going to ignore wage inequality, and write as if all workers get the same wage.

The simplest way to get to a universal basic income would be to pay it to everyone, then recoup the cost, through the tax system, from everyone above the basic level. While conceptually simple, this way of doing things would be almost impossible to implement except as a ‘big bang’, and is also too hard for me to evaluate. Instead, I’m going to look at a program that first raises existing basic incomes above the poverty line, then makes access to the basic income unconditional for those with no other source of income.

Initially, I assume that 20 per cent of the population is receiving a tax-funded basic income (old age pension, unemployment benefits and so on). Of the remaining 20 per cent, I assume that 10 per cent are retired and living on the income from savings (part of the capital share) while the other 10 per cent are dependents of workers, including non-employed spouses, students and so on, or else are getting by somehow in the informal economy. To keep things simple, I’ll treat them all as dependents of workers.

The set up so far means that the average worker earns income equal to average total income per (adult) person, and after tax income equal to 70 per cent of average total income. Allowing for dependents, each person in an employed household gets an average income equal to 60 per cent of average total income. (For the US, these figures are about $60k, $40k and $35k respectively).

Suppose to start with that the benefit is equal to 20 per cent of average earnings. That implies a cost equal to 4 per cent of total income. I’ll assume that a benefit equal to 30 per cent of average earnings is enough to put a person above the poverty line, which would raise the cost to 6 per cent of total income (the implied numbers are way above the official US poverty line – 12k and 18k per individual). Obviously, that’s not a big deal in itself – plenty of countries have conditional benefits that put nearly everyone above this (relatively tightly defined) poverty line.

So, what matters is making the benefit unconditional for those with no other income of their own. As a first step, that would mean paying the basic income to the 10 per cent of the population I’ve classed as workers dependents, raising the total cost to 9 per cent of income. That requires an increase in the total tax share from 30 to 35 per cent, which is significant, but still well within the range of ordinary variations. Assuming the tax increase is borne equally by labor and capital, the result is to reduce average post-tax earnings to 65 per cent of total income. However, because most of the benefits flow to workers’ dependents, there’s actually a small increase in the average income of worker households, net of taxes and transfers.

The big question is whether current workers will respond by leaving the workforce and relying on the basic income. We’d expect and want this to happen to some extent – the whole idea is to free workers from absolute dependence on wage income. But if the shift is too large, the tax burden will become unsustainable.

How large is too large? Suppose the employment rate falls from 60 per cent to 50 per cent, and that capital income falls in line with labor income, so that a larger benefit cost is being supported by a smaller income. The cost of benefits is now equal to about 15 per cent of total income, an increase of 10 percentage points from the initial position. Again assuming the cost is shared equally between capital and labour, the average tax rate would now be about 40 per cent. Depending on the design of the tax scales and the mix between income and other taxes, the marginal rate for the average worker would probably be around 40 per cent, and with a moderately progressive tax scale, lots of workers would be paying marginal rates above 50 per cent.

Summing up the exercise, I’d say that a universal basic income of the type I’ve sketched out here is economically feasible, but not, in the current environment, politically sustainable. However, while economic feasibility is largely a matter of arithmetic, and therefore resistant to change, political sustainability is more mutable, and depends critically on the distributional questions I’ve elided so far. A shift of 10 per cent of national income away from working households might seem inconceivable, but of course that’s precisely what’s happened in the US over the last twenty or thirty years, except that the beneficiaries have not been the poor but the top 1 per cent. So, if that money were clawed back by the state, it could fund a UBI at no additional cost to the 99 per cent.

More generally, political feasibility depends on both the path by which a policy goal is approached and the political and social framing of the issues. I plan to come back to this later, but I’ll be interested to read people’s thoughts now. Feel free to comment both on the validity of my estimates and on the political and social aspects – both will be valuable to me.

Note I’m not sure that this is the right framework in which to analyse a Universal Basic Income, in which everyone, regardless of means, would receive the payment, financed by taxation on those receiving more than the UBI, but I’ll have a go. I’ll look at a UBI that would yield final incomes, net of taxes and transfers, similar to those of the guaranteed minimum I’ve considered. In the framework of the example, the gross cost would be equal to 30 per cent of total income. Deducting the costs of existing welfare payments that would be replaced would bring this down to 26 per cent, but taking account of people shifting from employment to the basic income would raise it again, to a bit above 30 per cent. That would imply average tax rates above 60 per cent, and with modest progressivity, marginal rates of 80 per cent for workers on above-average incomes. That’s not inconceivable (existing systems of means-tested benefits cab have effective marginal rates of taxation above 80 per cent), but it seems a lot less feasible than the roughly equivalent GMI, which would be hard enough it itself.