Thursday, 31 March 2016

Strangling the UBI at birth

Labour's proposal for a UBI debate was swiftly scorned by Eric Crampton whose pithy summary runs like this:
There's a lot to like about a guaranteed annual income. Or, at least, there would be if it were feasible and affordable. I don't think it can be both.
This sounds both reasonable and devastating, right?  The claim is based on an "impossible trinity", which is that you can only pick two of:
* low phaseout rate
* Big basic benefit
* same cost as current system
From which, just as the man said: a UBI can't be both feasible and affordable. Don't bother yourself with defining these terms: just focus on the fact they sound like clear & reasonable hurdles: you either pass or fail. Much the same story is told by Jim Rose in his Taxpayers Union report, the Herald and Liam Hehir who deserves special mention for arguing we shouldn't even talk about it: the whole thing is impossible madness.

These are brave attempts to shut down the debate: strangle the whole idea at birth while demonising those with the temerity to suggest it. If that was your goal, you'd make your strawperson as extreme and narrow as possible.

So let's review the approach of Eric and his mates. At present, some people are net taxpayers and some are net recipients of welfare. Line up all these people in order from the person with the biggest net tax bill on the left of the following diagram, through to the largest net recipient of welfare at the right end.
To keep it simple, lets just focus on tax & welfare and pretend the red line shows the net position for each resident. In this case, the government's books are in balance for a year if the area A is the same size as the area B: all of the tax revenue is paid out in welfare. Because of mumble mumble, we need to make sure none of these residents is worse off, so the UBI has to be set at the net amount received by the biggest recipient, so under a UBI the total payout is equal to the combined area of B and C. Any twit can see that B + C > A, so there you go: utter lunacy that could only have been dreamed up by deluded fools who can't count.

Here are a few things they're missing.

  1. A more modest UBI might in fact take a lot of people out of the poverty trap created by targeting multiple welfare payments all of which abate as you earn extra income. Not the people at the far right of the diagram, but some.
  2. The government would save money by not paying people to monitor, grill, and generally hassle these citizens.
  3. Contrary to Eric's view, NZ's tax code does in fact have "have free lunches baked into it" and removing these would add to government revenue. The two most obvious opportunities are clawing back some of the Apple/Google/Facebook/... avoidance and taxing capital tied up in residential property.
My point is that there is actually a debate to be had, and it is very poor form indeed to deny that using strawpeople that do not withstand even casual scrutiny.


  1. Agree with your points 1 & 2 and say as much in my article. I then argue that a low UBI would result in a benefits system layered over the top of a UBI. You then get much of point 2 eroded.

    While I disagree with you on 3, let's grant the premise for now. If those are there, then you first fix the tax system so that it collects any amount of revenue at lowest deadweight cost. If the marginal cost of an additional dollar of spending is then lower, which it would be in this case, optimal government spending increases if ex ante expenditures were optimal ex ante. It would be... remarkable... if efficiencies in tax thus gained flipped UBI from being undesirable to being desirable

  2. I'm glad we agree there are design & evaluation issues on 1 & 2.

    You insist on ignoring complementary tax policy changes, preferring a perfectly optimal policy regime.

    Politically and economically, the only feasible path to a UBI must include complementary tax policy.

    So if you want to give the UBI due consideration, you need to embed it in politically feasible bundles of policy.

  3. You're putting a thumb on the scales then and should at least be explicit about it. If the loopholes you think exist are worth closing, then close them. If that means the marginal cost of raising money is then lower, and a UBI then passes muster *by more than do other potential uses of that money*, then fair enough.

    Agree that there's no feasible path to UBI without substantial tax hikes. But unless there's no chance that policies other than a UBI might be the best use of that increase in tax revenue, bundling them together seems more a political move than an economics one. Unless of course there are complementarities in the stricter sense: if the tax changes you prefer would be substantially different in effect under a UBI than they would otherwise, then I could see a case for bundling: in that case the value of the bundle would exceed the value of the two evaluated separately against the status quo. But I've not seen that case made.

  4. One option you ignore and have helped to foreclose is that (a) the "substantial tax hikes" are efficient for NZ but (b) these are most likely to succeed politically inside a broader package that includes a UBI.

    Also, as requested on twitter, can you explain please why you think that "Labour is right that the UBI is an idea worth exploring"?

    1. Worth thinking about for two reasons.

      First, if there is a way of doing it that could work, then that needs fleshing out. Like, I can imagine a version that combines a low UBI with a ramped up tax credit for donations to charities that fill remaining gaps, and maybe that could work, and seeing if that could work seems useful.

      That's the positive aspect of the debate.

      The other side of it is that we need to get rid of the wishful thinking versions of UBI.

  5. Thanks Eric. I'm pleased you are actually interested in the topic but disappointed you helped to strangle it at birth.

  6. hello, my name is Ngozi, a student of University of Nigeria. i find your work very educative and highly recommended for students worldwide. Please contact us at Thank you very much.