In principle, a Universal Basic Income, as floated by the NZ Labour Party, sounds great. It’s once you start looking harder at implementation that things quickly become, well, messy, writes Eric Crampton.

If you like a UBI, economist Kevin Milligan tells us you can choose two of the following three options. But only two.

1. A high enough basic income that few people on current benefits are made worse off;

2. A low phase-out rate so that lower income workers do not face sharp penalties for accepting work;

3. A cost that isn’t massively higher than current spending on benefit programmes.

You cannot have more than two of these.

And, it really is not hard to see why.

Universal Basic Incomes are universal. That means everyone receives the payment, regardless of income. While that means that government support is then not targeted to those most in need, it also can have rather a few advantages.

But benefit programmes, including Working For Families, are income-contingent. The amount you are paid depends on your household income – they phase out as your market income increases. This helps the government to target programmes to those in most need. But it also means that lower income earners receiving benefits can wind up with very little in the pocket for additional hours worked.Currently, beneficiaries receiving income support through multiple programmes can face very high effective tax rates on earnings. I do not mean the 10.5% or 17.5% income tax rates that apply on earnings up to $48,000 per year. Those tax rates are very low in the grand scheme of things.

How does that work? Suppose that your family’s taxable income were $20,000 per year. An additional dollar earned would see $0.175 sent off to the government in tax, if the family’s income comes from a sole earner. Fair enough.

But Working for Families and other benefits are also scaled back with that extra dollar earned. And the Tax Working Group showed that, in 2010, families earning just over $20,000 saw every extra dollar earned clawed back through the combination of benefit reductions and tax. It has improved a bit since then, but the problem remains substantial.

This builds what economists call poverty traps: ranges of income where, at least in the short run, additional hours worked or salary increases really do not leave the worker any better off. It can be hard to see the point of working harder if busting your hump for a raise just means you get a cut in other benefits that matches your pay increase. In the long run, those workers would be much better off by getting onto higher salary paths. But the short term incentives are very poor.

And so moving from the current set of overlapping benefit programmes to a single UBI has a lot going for it. The current mess of clawback rates could be folded into a progressive set of tax rates that do not create poverty traps. Those in need would not have to face the bureaucratic hurdles inherent in the current system. But it would come at a cost.

Beneficiaries with complex needs covered by a host of existing programmes receive substantial transfers from the government. For a UBI to really be universal, and to really avoid the poverty trap problems that result from overlapping income-contingent benefits, a UBI would need to pay enough to leave those beneficiaries no worse off.

But extending that level of payment to every adult in the country would be phenomenally expensive. Labour’s recent discussion document suggested the payment would need to be $22,000 per adult. Currently, the government collects $72 billion per year in tax. That covers everything from income transfers and benefits, to the police and defence. If the government paid $22,000 per adult as a guaranteed annual income, it could afford to pay that amount to just over 3.25 million people – and have nothing left over for anything else.

And so we come back to the trilemma with which we opened. A UBI that pays enough to leave current beneficiaries no worse off would be phenomenally expensive and would require substantial tax increases, if it were not combined with a very high clawback rate. But high clawback rates are part of the problem that a UBI is meant to solve.

Alternatively, Labour’s discussion document suggested combining a lower UBI with supplemental support packages for those in greater need. But layering a welfare system on top of a UBI gives us many of the costs of a UBI, with far fewer of the advantages. Those in need would still have to jump through hoops to receive additional support. If that support comes in form more like current WINZ special needs grants than like current programmes, it will be harder for those currently in most need to receive support.

And there really are no dollars lying around on sidewalks ready to be picked up by the government to fund a UBI. If changes to the tax system make sense, they make sense regardless of whether the government wants to use the revenue to fund a UBI, to fund other programmes, or to cut other tax rates.

In 2010, Treasury worked out what would happen if the government replaced all existing benefits, including NZ Super and Working for Families, with a universal payment of $300 per week for adults plus $86 per child per week. They found that income tax rates would have to rise to over 55% to fund the scheme. And, the proportion of people living on less than 50% of median household disposable income would rise by 5%.

The system would be costly and would leave the worst off worse off. If the worst off are left worse off, political pressure for layering welfare programmes on top of a basic income scheme would not be small. And hiking tax rates from 33% to 55% would be rather damaging. This was not the tax rate that Treasury thought would apply only to rich people – it would apply to everyone.

Milligan’s impossible trinity cannot be avoided.

Proposals like Gareth Morgan’s Big Kahuna scheme try to square the circle by combining relatively low basic benefit levels with a new comprehensive tax on capital. The scheme would tax any owned capital as though it had earned a 6% return, then tax that return at 30%. So your $500,000 house would cost $9,000 per year in capital tax, regardless of your income or ability to pay.

But the scheme suffers from two bigger problems.

First, if a comprehensive capital tax makes sense, then it makes sense regardless of whether there is a UBI. It should be evaluated on its own merits. For a start, assuming that all capital earns a 6% rate of return, and taxing that assumed return, seems more than a little harsh in a world where term deposit rates are closer to 3%.

More importantly, though, while relatively low basic benefits could make the system affordable, they would not be politically stable. There would be pressure to layer a benefits system on top of the UBI, or to increase the UBI. It does not seem plausible that any government would be able to withstand the likely months’ long John Campbell campaign that would, every day, highlight a different family whose benefits were cut under the shift to a UBI. We would quickly have a welfare system layered on top of a UBI, increasing the costs while eroding the UBI’s benefits.

If all of that were not reason enough to worry about UBI proposals, think through also the implications for immigration policy. Safety net programmes encourage governments to build walls and moats around those programmes to help keep costs down.

During last year’s refugee crisis, the government balked at increasing the refugee quota, noting the costs of refugees to the health, welfare and education systems. If every new migrant to New Zealand came with a hefty expected UBI invoice attached, New Zealand would accept fewer migrants. If permanent residents were eligible for the UBI, we might well expect that a government wishing to keep costs down would accept fewer new permanent residents – and fewer migrants who might risk becoming permanent residents.

Do we really want to turn people away at the border, people whose lives would be better here and who would help to make our lives better in turn, because they could be too costly to the public purse through a UBI?

I love the idea of universal basic income schemes. Simply giving cash to poor people should always be the baseline against which every other government intervention programme should be assessed. If some programme meant to make poor people better off does them less good than simply giving cash, that programme should be recognised as a failure. And sorting out the poverty traps created by our current tax and transfer systems is really important.

But there really does not seem to be a universal basic income that can solve those problems without creating worse ones. Labour is right that the UBI is an idea worth exploring, but do not get your hopes up that anyone will be striking gold here.