Tuesday, June 7, 2016

Universal Basic Income

Universal Basic Income is in the news. Charles Murray wrote a thoughtful piece in the Wall Street Journal Saturday Review. The Swiss overwhelmingly rejected a referendum -- but on a proposal quite different from Murray's.

Murray proposes that "every American citizen age 21 and older would get" $10,000 per year "deposited electronically into a bank account in monthly installments." along with essentially a $3,000 per year health insurance voucher.

The most important part of Murray's proposal: UBI completely replaces
 Social Security, Medicare, Medicaid, food stamps, Supplemental Security Income, housing subsidies, welfare for single women and every other kind of welfare and social-services program, as well as agricultural subsidies and corporate welfare. 
There is a lot to commend this idea. First, it would reduce the dramatic waste in the current system:
Under my UBI plan, the entire bureaucratic apparatus of government social workers would disappear
Moreover, the bulk of government spending now does not go to people who are really poor. SSI and medicare go to old people, many of whom are quite well off. Housing subsidies such as the mortgage interest deduction go to people with big mortgages and big tax rates -- nor poor people. Murray doesn't really emphasize this point, but his proposal is far more progressive than the current transfer system.

Second, it would reduce the very high disincentives of the current system, which traps people.
 Under the current system, taking a job makes you ineligible for many welfare benefits or makes them subject to extremely high marginal tax rates. Under my version of the UBI, taking a job is pure profit with no downside until you reach $30,000—at which point you’re bringing home way too much ($40,000 net) to be deterred from work by the imposition of a surtax.

If I read Murray correctly, he takes away $3,500 of the benefit between $30,000 and $60,000, which is an 11.6% surtax. That applies on top of the Federal 25% marginal rate, 16% payroll tax, state income and payroll taxes and so forth. So not zero, but it is a lot less disincentive than many current programs.

Both considerations place the proposal not in the "perfect world" category, but "how can we do what we're trying to do now a lot more effectively." So, evaluate it as such.

The biggest problem in the argument is the biggest selling point: We trade a check -- even much more than $10,000 -- for complete elimination of everything else.
A UBI will do the good things I claim only if it replaces all other transfer payments and the bureaucracies that oversee them. If the guaranteed income is an add-on to the existing system, it will be as destructive as its critics fear.
There are a lot of these "big trades" on the table, and there should be more. A big carbon tax, in return for complete elimination of all the regulatory nudges and crony energy related subsidies. A VAT in return for complete elimination of income, corporate, estate, and other taxes.  Lots of infrastructure money in return for elimination of Davis-Bacon, endless legal challenges EPA reviews, and other regulations, strict cost-benefit analysis rather than subsidized anachronisms, and so on.

In all these much simpler cases, the deal doesn't get off the ground. Will the "right" allow a big enough carbon tax? Will the "left" really get rid of their subsidies? Will the "right" really allow a large enough VAT? Will the "left" really not just pile all the other taxes back on top? Making these deals is hard enough even when both sides admit the deal would be good.

That case is going to be even harder here. The "left" has not even thought about the deal, let alone agreed in principle with only trust issues remaining! The Swiss referendum [sad aside on media: it was really hard to find the actual text!] made no mention at all of a swap -- it was pure basic income on top of other social programs.

Programs will remain tempting, because a flat basic income is not close to the "perfect world" social insurance system, or even common sense. We want to give more help to people who need more help. That lets us be more generous to those who do need help, and contains moral hazard that people who don't really need help should be working and paying taxes to supply help. Social security goes to old people, because old people objectively are less able to work.  Disability goes to disabled people, because it's harder for them to work as well. Unemployment insurance goes to people who just lost jobs, we know they are more likely to have suffered a bad shock. Insurance payments go to people whose houses have burned down.

These social insurance programs are indeed ineffective, bureaucratically bloated, and do a terrible job of picking who really needs help from who doesn't. But UBI takes a pretty extreme view that the project is completely hopeless, and the Government should do no conditioning at all, other than reported income:
Government agencies are the worst of all mechanisms for dealing with human needs. They are necessarily bound by rules applied uniformly to people who have the same problems on paper but who will respond differently to different forms of help.
Well, ok, but the call of the better world will be hard to resist, and the "left" has far from accepted that bureaucracies are "the worst" mechanism for sorting the needy from the less needy.

There will still be unfortunate people,  they will still need help, and our electorate will still demand programs to help them. Disability: Ok, it's grown  out of control, but some people really are disabled. You're only going to give them $10,000 and turn your back? What about the guy who takes his check, blows it all on a weekend of meth and beer, and now is lying in the gutter, his children homeless?
Some people will still behave irresponsibly and be in need before that deposit arrives, but the UBI will radically change the social framework within which they seek help: Everybody will know that everybody else has an income stream. It will be possible to say to the irresponsible what can’t be said now: “We won’t let you starve before you get your next deposit, but it’s time for you to get your act together. Don’t try to tell us you’re helpless, because we know you aren’t.”
He goes on to extol the virtues of private charities. I don't think our electorate is ready to completely forswear all bureaucratic help. And the vine grows back.

Eliminating housing subsidies? Agricultural subsidies? "Corporate welfare?" These are all great ideas on their own. If we could do that, our economy would be in a lot better shape than it is.

A bit of paternalism is pretty ingrained in social policies, and it isn't necessarily a bad thing. I'm happier paying taxes to support food, clothes and school for the kids, and basic housing than I am to subsidize a beer and meth weekend. Murray already gives in, by restricting the first $3,000 to a health insurance voucher. If he's going to get rid of social security, he should restrict the next $1,000 to a forced savings plan. If we're going to get rid of all housing programs (a great idea) the next $2,000 is a rent/mortgage voucher.

Some paternalism is justified as a pre-commitment. We know if they blow the money, we'll enact social programs to help them after the fact.

There is a deeper problem -- and I have a constructive solution.

In fact, Americans use far fewer benefits than they are eligible for. Many programs have 2% take up rates. Lots of people eligible for medicare, Obamacare subsidies, disability food stamps, welfare, home heating subsidies, and so on and so on all the way down to Palo Alto's income-based parking permit system don't take advantage of the benefits. If each American took advantage of every subsidy and social program to which he or she is entitled, the country would be bankrupt in about 10 minutes.

Why not? Well filling out the forms is a pain. And, more importantly, most people really do use social programs for a limited time. Call it a stubborn independence ethic or some remaining shame to taking assistance, it's there. For now. I fear that welfare states fall apart when the social stigma of taking the money fades.  

For now, both act to limit moral hazard. If it takes a few hours and trips down to an unpleasant bureaucracy to get help, then only people who really need it are likely to ask. If there is some remaining social stigma to getting help, then only people who really need it are likely to ask -- and likely to get out as fast as possible.

Before I get howls of comments on how heartless this view is, remember the objective -- money is limited, we want to use it to help people who really need it, and if we can do something to keep out people who don't, we can be a lot more generous to those who do. If we impose some cost on people to get help, we get them to reveal who really needs it, and we can help them a lot more.

So, my major suggestion -- please, don't automatically send the check to every American the minute they turn 21! Don't send it to my kids! At least, make people go down to a dull and dirty office, stand in line, fill out a long form, and repeat once a year.

Murray limits the benefit once you get to $30,000 per year, introducing a surtax above that level. I've been mulling over a different way to limit benefits and thereby make them more generous: Limit by time, not by income. You can have an additional (say) $10,000 per year, for 5 years, at any point in your life. Most people using social programs do in fact use them to get out of trouble and back on track. Let's make that the expectation. This is not permanent income support, this is help to get out of trouble.  That lets us be more generous, without blowing the budget, and without inducing as large a marginal tax rate to working.

Murray has a lot of speculation on how society will adapt to $10,000 per year check and NO other social programs.
the entire bureaucratic apparatus of government social workers would disappear, but Americans would still possess their historic sympathy and social concern. And the wealth in private hands would be greater than ever before. It is no pipe dream to imagine the restoration, on an unprecedented scale, of a great American tradition of voluntary efforts to meet human needs. 
Trust private charity, with an ever-larger share of income in plutocratic hands? I don't see Bernie Sanders supporters signing on to the deal on that basis.
The known presence of an income stream would transform a wide range of social and personal interactions. The unemployed guy living with his girlfriend will be told that he has to start paying part of the rent or move out, changing the dynamics of their relationship for the better. The guy who does have a low-income job can think about marriage differently if his new family’s income will be at least $35,000 a year instead of just his own earned $15,000.
Or consider the unemployed young man who fathers a child.
Maybe. Maybe not. We do have some experience with corners of societies that live off government checks. We have more experience with places where lots of people don't work. Welfare neighborhoods in the 70s to mid-90s. Europeans living on the dole. Molenbeek. Saudi Arabia. By and large, places where most people live on government checks or large numbers don't work are not happy places.

One can also speculate in contrary ways. Labor markets are more and more regulated and restricted. Well, if people can all get $10,000 from the government, why fight for lower minimum wages for entry level workers, looser occupational restrictions, and so forth?

Murray also confuses the issue, and substantially weakens the case, I think, by wandering off into a soliloquy on once robots do everything there won't be any more jobs.
We are approaching a labor market in which entire trades and professions will be mere shadows of what they once were... the jobs (now numbering 4 million) that taxi drivers and truck drivers will lose when driverless vehicles take over... Advances in 3-D printing and “contour craft” technology will put at risk the jobs of many of the 14 million people now employed in production and construction...The list goes on, and it also includes millions of white-collar jobs formerly thought to be safe..
... as many as 47% of American jobs are at risk...it will need to be possible, within a few decades, for a life well lived in the U.S. not to involve a job as traditionally defined.  
I think this is wrong. Murray acknowledges

I’m familiar with the retort: People have been worried about technology destroying jobs since the Luddites, and they have always been wrong.

Indeed they have. The invention of the tractor was way worse than the invention of the self-driving car for the jobs of about 70% of Americans and about 99% of everybody else at the turn of the 20th century -- farm labor. Murray writes
 It takes a better imagination than mine to come up with new blue-collar occupations that will replace more than a fraction of the jobs..
It's a good thing that every time in the past we did not rely on policy writers' imaginations to come up with occupations for people. I think the answer is pretty clear: services. When robots make everything for us, then people make money supplying services to each other.

But I don't have to be right either.  The deeper problem with this line of argument, common on the left, is how utterly hopeless it is, and how it contradicts Murray's case.

Hopeless: Really? Your vision for the future is that 47% of working-age Americans will be living on a $10,000 per year check from the government, doing nothing? $10,000 is not a lot of money, barely sustaining a life on the margins in pockets of poor rural america. It buys a used trailer and a six pack of beer in a place with little hope.

We can do better than that! And we can. We're talking about a several decade shift in the labor force here. If services are the answer, we need to fix schools and other barriers that keep people from getting the skills needed to earn money in the service economy. We need to fix labor markets to make it easier to hire people in flexible ways and help them to develop skills on the job.

Contradictory: Murray's numbers work out (I think, I haven't checked, but it seems plausible) in today's America. But if half our labor force, and all our retired or non-working people, are living off a government check, the cost would explode past what the country could possibly support with any level of taxation.

So set this apart, recognize that adapting to automation will require getting people skills not sending them checks. And that is going to mean keeping the price system alive. It has to be crystal clear that computer programming pays more than goof off majors.

Bottom line, most of the Murray's social changes and adaptation to robot workforce is, I think, a mistake and a distraction.

A Big Deal -- along with the others -- remains attractive: Substantial cash grants and vouchers in place of many current programs -- could offer substantially more help to people who need it, with far fewer distortions.  In place of middle class subsidies -- housing, college, etc. -- and corporate subsidies even better.  But let's not pretend it will cure social ills, or save us from confronting labor market distortions.


  1. So won't we have entire political campaigns/parties devoted to raising the UBI significantly above inflation?

    1. I can already picture the calls for a "Living UBI."

    2. Pro life if you will

  2. This seems to be the full text https://www.parlament.ch/it/ratsbetrieb/suche-curia-vista/legislaturrueckblick?AffairId=20140058 of the Swiss initiative, chronology here https://www.parlament.ch/it/ratsbetrieb/suche-curia-vista/geschaeft?AffairId=20140058 (this certainly does not mean I know Italian, but a bit of German helps).

  3. Not a comment on UBI, but I believe that your farm/tractor analogy is less than apt because the replacement for those farm jobs did not require an education: move to an urban center and work in a factory.

    That scenario no longer applies; you cannot leave a service industry or manufacturing job that has been automated out of existence and expect to provide the new, "needed" service that will keep you gainfully employed. Especially, teaching.

    I agree that a multi-decade transition is required, but it is getting from A to B that is worrisome.

  4. I can already see an advantage to this for macroeconomic stabilization purposes. Index the UBI to the CPI (or the PCE price index). If the central bank adopts a price level target, it would be easy to communicate the PLT to the public, which currently seems to be an obstacle to adopting a PLT. Individuals can form expectations based on expected UBI at date t+i, which is indexed to the PLT.

    1. A big problem with indexing either the UBI or the minimum wage is that it cause a vicious cycle.

      indexed minimum wage --> prices of low-skill-labor-intensive goods --> minimum wage

      indexed UBI --> reserve price of low-skilled labor --> prices of low-skill-labor-intensive goods --> UBI


  5. Implementing UBI to cure the social pathologies created by the welfare state is sort of like taking cyanide to cure your cancer. If you take enough cyanide, you won't be suffering from cancer anymore, but ...

    I think he most effective policy combination would be to completely legalize and deregulate drugs while implementing UBI. Within 5 years, the OD deaths will eliminate unemployment.

  6. Restricting thought to a single state solution for a basic income is a disservice to mankind, and counterproductive to world peace and stability.

    A structure to provide a global basic income to every adult human on the planet could be created and maintained for the same cost, or less, as a single state program.


  7. I agree that a multi-decade transition is required, but it is getting from A to B that is worrisome.

  8. The latest print issue of the The Economist (from June 4) has a 4-page "Briefing" on the same topic, and a related editorial as well.
    One point on the Swiss referendum: my understanding is that the proposal in front of voters was *an addition* to the current welfare programs already in existence in Switzerland. Murray's proposal is a substitution.

    1. http://www.economist.com/news/briefing/21699910-arguments-state-stipend-payable-all-citizens-are-being-heard-more-widely-sighing

  9. What if no one will sell you health insurance for $3,000 per year?

  10. I have a similar idea in another area.

    Bankrobbers cause a deadweight loss in hiring guards and enforcing protections. That loss is probably greater than the money now stolen. I propose that all bankrobbers be given a $10K support distribution in exchange for their agreement to stop robbing banks.

    It might be too easy for a person to claim that they were a bankrobber and get an undeserved benefit. So, always make these people go down to a dull and dirty office, stand in line, fill out a long form, and repeat this once a year. People who were not bankrobbers will find this somewhat demeaning, and it will limit the amount of undeserved handouts.

    OK, this is a sloppy analogy. Poor people inspire more sympathy than bankrobbers. But, other parts of the analogy organize this argument.

    The current welfare system did not come about through mistakes. Politicians have put it together step by step to buy votes and hire multitudes of welfare bureaucrats. Bootleggers and Baptists are always with us. Politicians (bootleggers) benefit themselves under the cover of a moral purpose supported by the Baptists.

    === ===
    Cochrane above: [Why don't people apply for all the benefits now available?] Well filling out the forms is a pain. And, more importantly, most people really do use social programs for a limited time. Call it a stubborn independence ethic or some remaining shame to taking assistance, it's there. For now. I fear that welfare states fall apart when the social stigma of taking the money fades.
    === ===

    The best way to take the shame out of assistance is to make it universal, automatic, just for filling out a form. It is a mystery that more people are not applying for welfare. The bureaucracy must be so convoluted and incompetent that many poor people doubt that they would complete the paper requirements.

    The big flaw with attempting to trade simple, automatic assistance for dismantling the current labrinth is that there is no one reliable to bargain with. Five years after making the grand bargain with current politicians, future ones will reassemble the current mess step by step under the same incentives which motivate current policies.

    The only answer to a bad system is to dismantle it in principle, not to make it temporarily more efficient.


    Milton Friedman: "I have no apologies for it, but I really wish we hadn't found it necessary and I wish there were some way of abolishing withholding now."

    === ===
    At the link: By making taxes relatively invisible and making paying them relatively painless, Friedman paved the way for the current status quo, in which significant portions of salaried workers' incomes are handed over to state and federal governments without much fuss. Tax refunds, which feel like a delightful windfall to those who receive them, complicate the matter further by making many Americans feel like they are getting a gift on tax day, rather than giving their hard-earned money to the government under threat of violence.
    === ===

    The above proposal for automatic, universal assistance would further pave the way toward a deep socialism and eventually the Greek or Venezuelan model. Every move toward making a system efficient gives approval to that system. This doesn't bother current politicians. Almost all of them take the 18th century French royal view, "After me, the flood".

  11. I favor a "fulcrum tax" which addresses some of the issues John raises. All of us benefit from the accumulated wisdom and knowledge of our species. I would argue that the dividend from this accumulated capital should be distributed equally across society. I would also argue that the gains and losses of creative destruction are highly skewed -- e.g., globalization has had huge benefits for the world's poor and US consumers, but has devastated workers in the US manufacturing sector. They are also unpredictable and can hit anyone at any time -- e.g., the AI revolution could someday replace many of today's highly skilled employees.

    My proposed solution is a fulcrum tax. The idea is to lower everyone's beta to the economy by sharing in the (often unpredictable) gains and losses from creative destruction. It works like this: If you make more than the average income, you pay 25% of the difference to a redistribution pool; If you make less, you get 25% of the difference back.

    Some advantages of this approach:
    1. It can be done entirely by computers; no money flows through Washington, minimizing the potential for cronyism and corruption.
    2. It provides a minimal annual income for all Americans (25% of the national average)
    3. It preserves consumer choice -- no statists telling people how to spend the money (i.e., SNAP, Obamacare, etc.)
    4. It could make the social security trust fund solvent; the fund would only need to cover the difference between the calculated benefit and the minimum annual income.
    5. It would require recipients to file a tax return -- no benefits to illegals
    6. If the average income goes up but the median income is flat (i.e., the well off are getting most of the gains), everyone still benefits (so people will be happy when national income goes up, even if their personal income is flat)
    7. Other costs of government could then be covered by a small VAT or national sales tax, which is much easier to collect.
    8. Etc.

  12. Wouldn't 18 be a better age than 21? Many people are independent at 18. And those that are still dependent are likely in school. (Perhaps one more subsidized industry that can be wrapped up in this.)

  13. “For now, both act to limit moral hazard. If it takes a few hours and trips down to an unpleasant bureaucracy to get help, then only people who really need it are likely to ask.”

    In Canada, we see a form or rent emerging with the unpleasant bureaucracy for people with disabilities. Long term disability benefits are paid out of the Canada Pension Plan (CPP), but the processes and forms required for establishing a disability are labyrinthine. We now see disability specialists who run their business by helping people navigate the processes for a portion of the entitlement. Between the tax lawyers and accountants, the potential problem with an onerous bureaucracy is the rent it creates as people try to navigate the system.

    Can we not combine the best of UBI with a bit of conditioning?

  14. I recently talked to a social worker in Austria working with unemployed people. There are two classes of contribution levels for people out of work for longer Terms. Those still capable of work, at least theoretically, and those not for whatever reason. The first ones have to prove their principal readiness to work and get roughly 800$ and the second one get about 600$.

    The social worker told me about a guy at 24 in full health who never was in work and managed to get the higher rate due to his ability to "distract" potential employers. She told me from a similar guy who had made an ugly tatoo in his face to achieve a similar effect.

    So it is not necessary the fifty years old former mun who raised three Kids that is getting These benefits.

    The benefit levels are not high but for some people who value having no stress at all and always a good sleep every day it seems the better choice.

    I think you have to approach it not only from the standpoint what makes economically sense for the working majority but also which level of abuse is acceptable to Society and this increases with the Level you hand out.

    You obviously have decrasing marginal benefits from Basic income , as you have increasing abuse with Basic income rising. Finding the pareto Optimum is the challenge.

  15. I think your analogy of what the introduction of tractors did to farm work would be more similar to what robots will eventually do to current factory (and many services) work if seen from the point of view of horses, not humans. Hhmmmm... seems that didn't work out so well for horses, did it?
    I find more than a bit disingenuous to argue that services will always provide work for the victims of technological unemployment, so their lack of a salaried occupation can be assumed to be their fault only. "Services" sound great if you think of web programmers and financial consultants, but not so much if what you have in mind are janitors, toilet scrubbers and burger flippers (which is closer to what is realistically available for the vast majority of the lower middle classes). Will there always be work to be done in exchange for a pittance? yes, the work that the best educated do not want to do. Is such job a source of dignity and self-worth, especially if its remuneration is dictated by an unregulated market? I seriously doubt it. But I can see how the "knowledge workers" that are confident in their ability to keep on earning big bucks may oppose a UBI, it will keep the cost of their servants low...
    I normally enjoy your posts very much, and tend to agree with most of them, but I find this one widely off the mark (and find more congenial Murray's position, for reasons I expound on this post: http://purebarbell.blogspot.com.es/2015/03/the-conservative-case-for-ubi-what-lies.html after reaching similar conclusions about what would need to be defunded for it to be feasible: http://purebarbell.blogspot.com.es/2015/01/where-would-money-for-really-big-ubi.html)

  16. "$10,000 is not a lot of money, barely sustaining a life on the margins in pockets of poor rural America."

    Some USA citizens live in less than that now. Some even voluntarily see here: http://earlyretirementextreme.com/how-i-live-on-7000-per-year.html

  17. The recent campaign by Andrew Yang features a UBI proposal (https://www.yang2020.com/what-is-freedom-dividend-faq). It basically gives every American $1,000 per month, no strings attached. Could you please comment on it? I'd be very interested in learning your view. Thank you very much!


Comments are welcome. Keep it short, polite, and on topic.

Thanks to a few abusers I am now moderating comments. I welcome thoughtful disagreement. I will block comments with insulting or abusive language. I'm also blocking totally inane comments. Try to make some sense. I am much more likely to allow critical comments if you have the honesty and courage to use your real name.