The Universal Basic Income is only a tangential interest of mine. Yet when I’ve debated it, I’ve been consistently impressed by how little the eager advocates try to teach me.* Case in point: I learned more from reading three paragraphs in Kevin Lang’s Poverty and Discrimination than in my typical conversation with a UBI enthusiast:
Because the stakes involved in instituting a negative income tax were so high, policy analysts convinced the federal government to conduct experiments in which some people were randomly assigned to be eligible for the negative income tax while others were randomly assigned to remain subject to traditional welfare. In the experimental group, there was also variation in the generosity of the program. Four experiments were conducted in the United States and a fifth in Canada. The largest of these is known as SIME/DIME (the Seattle Income Maintenance Experiment / Denver Income Maintenance Experiment).
Many policy analysts found the results of the experiments disappointing. Although the labor supply response was modest, it added substantially to the cost of the program. Depending on the generosity of the program evaluated in SIME/DIME, the labor supply response could account for over half of the costs. The least generous program would save $4 billion but would make 95 percent of recipients worse off. A program that would guarantee support at the poverty level and tax-back benefits at a rate of only 50 percent would still make one-fourth of recipients worse off and would exceed the cost of the welfare program then current by $30 billion, an enormous increase.
There were also some “unintended consequences,” the social science equivalent of medical side effects. In particular, the divorce rate rose among recipients randomly assigned to the negative income tax. The combination of the costs of the labor supply effects and the effect on marriage led Senator Daniel Patrick Moynihan, an early supporter of the negative income tax, to withdraw his support.
If I were an enthusiastic UBI advocate, I would know this experimental evidence forwards and backwards. Almost all of the advocates I’ve encountered, in contrast, have little interest in numbers or past experience. What excites them is the “One Ring to Rule Them All” logic of the idea: “We get rid of everything else, and replace it with an elegant, gift-wrapped UBI.” For a policy salesman, this evasive approach makes sense: Slogans sell; numbers and history don’t. For a policy analyst, however, this evasive approach is negligence itself. If you scrutinize your policy ideas less cautiously than you read Amazon reviews for your next television, something is very wrong.
*Exception: Ed Dolan
READER COMMENTS
Miguel Madeira
Sep 5 2019 at 11:11am
What is the marginal tax rate in these negative income system?
Putting in other way – probably the biggest theoretical advantage of UBI against traditional welfare is that UBI, supposedly, does not have the big implicit marginal tax rates of traditional welfare (where, if you raise your income, you lost many – or all – of the subsidy); but, if a negative income tax has also a big marginal rate, is the same thing (only in a more direct way).
Floccina
Sep 5 2019 at 12:23pm
Democrats tell us that people will still work almost as long and hard up to a 70% tax rate. France actually taxes up to 55%, so I would think we would want to tax away an NIT at a 55% rate.
robc
Sep 5 2019 at 1:36pm
I back of enveloped one a year or so ago and ended with a flat income tax rate of between 35 and 40 percent with no deductions at all.
I cut government spending to make it happen pretty harshly too. Assuming static conditions (which is a bad assumption*), I balanced the budget and had everyone a UBI of about 10k per adult and 4k per child.
I oppose the UBI for a number of reasons, but the practical one is because I think that is about as high a rate as we could go and I don’t think the spending cuts are politically possible.
*For one thing, a lot of government workers would be unemployed and we wouldnt be getting their tax dollars. I think that over time, that would work out for the best, but there would be an immediate shock.
Arnold S Kling
Sep 5 2019 at 12:42pm
Those three paragraphs are not the end of the story. See
https://pdfs.semanticscholar.org/38d1/bbb31dc6090a741f0793b6b166f82f698cbe.pdf
A lot of the income of people close to the poverty line goes under-reported, and this complicates the evaluation of these sorts of experiments.
The essence of economic thinking is the belief that incentives matter. Substituting a welfare program (UBI) that has a relatively low marginal tax rate for welfare programs that have “cliffs” changes the incentives in favor of working. In this case, I go with economic reasoning over results from experiments that are not clean, that take place over a short period, and that are confounded by phenomena like under-reported income that may differ between experimental subjects and controls.
DeservingPorcupine
Sep 5 2019 at 3:26pm
Aren’t you forgetting that free money can cause incentive to work to drop to 0?
DeservingPorcupine
Sep 5 2019 at 3:16pm
One huge problem with these experiments is that they’re temporary. If I know I’m getting free money for a little while, I might choose to sit on my butt a bit, but I’m still going to stay pretty “warmed-up” to the job market by doing things like updating my skills and preventing too large of a gap on my employment history.
Know that I’d get free money for life, though, is radically different. These experiments are, at best, finding the floor for work disincentives.
robc
Sep 5 2019 at 4:48pm
A UBI-like plan that doesn’t suffer from that problem is the one suggested by Thomas Paine (yes, that one).
It is a one-time lump sum payment on the 18th birthday.
I don’t remember the amount of his plan, but I would suggest something around the median household income, lets say $60000.
This is done with the understanding that there would be absolutely NO welfare type payments coming in the future.
As someone said upon hearing this, there would be a great growth in industries to separate that money from an 18 year old.
But a lot of good could be done too. It would pay for 4 years at a reasonably priced university. Drop it all in an index fund and 50 years later its a nice start to a retirement. It will also buy a lot of blow. YMMV.
Christophe Biocca
Sep 5 2019 at 6:43pm
The GiveDirectly people are pretty much A/B testing this right now in Kenya.
There’s an experimental group receiving monthly UBI for 12 years, another getting the same payments for 2 years, and another getting the NPV up front as a lump sum. Plus a control group.
The recipients (and the amounts) are very different from anything that’s being discussed in OECD countries but it’s something to pay attention to.
SomeDude
Sep 5 2019 at 3:39pm
Is anybody aware of a comprehensive review of UBI experiement resutls?
Todd Kreider
Sep 5 2019 at 3:51pm
A 1990 University of Wisconsin study argued that the divorce rate did not go up among those receiving the negative income tax relative to the control group:
Abstract: … … The reanalysis of the experimental data distinguishes between the experimental treatment in the form of the “pure” negative income tax and the treatment plans that involved an experimental training program. All the time periods of the experiment are used, with allowance for the timing of the marital dissolution in the inferences, and with allowance for attrition. The conclusion of this paper is that the plans (specifically, the negative income tax plans in the experiment) had no effect on the rate of marital dissolutions among the “treatment” couples relative to the control couples.
https://www.jstor.org/stable/2780516?seq=1#page_scan_tab_contents
Chris
Sep 5 2019 at 5:10pm
Just noting that an increase in divorce rates is not necessarily a bad thing, depending on what is leading to the divorces. If the UBI happens to be increasing the independence of both parties in the relationship to a point where they are no longer financially reliant on the other, then that’s actually a huge success. If it’s creating increased economic stress in some way and that is leading to the divorces, that would be negative.
Thaomas
Sep 5 2019 at 6:17pm
I’d like to see an analysis/experiment that bears on a UBI compared to an EITC. Yes, that says nothing about UBU v other forms of welfare, but it does/could make recipients ineligible on income grounds for various means-tested benefits.
Chris
Sep 5 2019 at 6:38pm
The paper linked to by Arnold is a helpful resource and definitely worth reading for a better idea of what different people found through analysis, as well as the significant limits of the data collected in these experiments.
For instance, the reduction in labor supply in young people was almost completely accounted for by an increase in school attendance in one test; people appeared to be reducing work hours to attain better education.
Another item of note is that the experiments didn’t show any increase in health of participants, or increase in community participation, but did show a huge increase in migration correlated with the UBI, which is another big positive as the ability to move leads to significant economic advantages.
Mark Z
Sep 6 2019 at 1:59am
I don’t think we should take it for granted that either education or migration is a net positive here; either could just as easily be a form of consumption. If education is mostly signaling, then it may be unlikely those pursuing education are increasing their productivity enough to justify the cost; and if someone migrates because they can collect the check anywhere, it can lead them somewhere where they are less productive as easily as somewhere they are more productive.
Comments are closed.