It’s not economically inefficient for a UBI to reduce recipient’s employment

A UBI (e.g. paying every adult American $8k/year) would reduce recipient’s need for money and so may reduce their incentive to work. This is frequently offered as an argument against a UBI (or as an argument for alternative policies like the EITC that directly incentivize work).

This argument is sometimes presented as economically hard-headed realism. But as far as I can tell, there’s not really any efficiency argument here—there’s nothing particularly efficient about people having a stronger incentive to work because they are poorer. The argument seems to mostly get its punch from a vague sense that work is virtuous and necessary. I think that sense is largely mistaken and it should be taken less seriously.

(As usual for policy posts, I’m not very confident about any of this; if it makes you happier feel free to imagine sprinkling “seems to me” throughout the prose.)

What people fear

If I give you $8k, you will probably value marginal dollars less than you used to. Some jobs that used to be a good deal will stop being worth it, and a job search itself may stop being worthwhile. We could study how much this happens empirically, but it’s definitely plausible on paper and it would be my best guess for the long-run effect even if pilot UBI experiments found otherwise.

This seems to be one of the main worries people have about a UBI. For example, some folks at McKinsey write:

In the design of the Finnish experiment, the main research question, agreed to by parliament in the enabling legislation, was the impact of a basic income on employment. Many policy makers assume that an entirely unconditional guaranteed income would reduce incentives to work. After all, the argument goes, why bother with a job if you can have a decent life without one? This assumption has led many countries to deploy active labor-market policies that require people on unemployment benefits to prove their eligibility continually and, often, to participate in some kind of training or to accept jobs offered to them.

[They find that the UBI didn’t decrease work, but this post isn’t even going to get into that.]

But it’s left somewhat vague what exactly is so bad about this.  In fact it’s hard to respond to this concern because, although I’ve seen it expressed so many times, I’ve never really seen the argument laid out clearly.

It’s clear that work creates much of the great stuff in our society, and so reducing how much work happens seems scary and bad. But when I work I get paid, so that I (and the people I share my income with) get most of the benefits from my work. And leisure and slack also create value. If it stops being personally worth it for me to work, then it has likely stopped being socially efficient for me to work, and that’s OK.

The real cost of a UBI is on the taxpayer’s side, and so the actual focus of discussion should be “how expensive is a UBI and are we willing to pay that much?” Thinking about recipients’ incentives to work on top of that is double-counting at best. To be clear, I think that the cost of a UBI usually is the focus, as it should be. So all I’m trying to do is address a little bit of FUD about UBI rather than make the positive case.

My view

Is working good for the rest of society?

Suppose you do some work and earn $100. The question from the rest of society’s perspective is whether we got more benefit than the $100 we paid you.

We can get more than $100 if e.g. you spend your $100 on a Netflix subscription that subsidizes better TV, or help our society learn by doing and advance technology for the products you produce or consume.

We can get less than $100 if e.g. you spend your $100 renting an apartment in a city that crowds out others, or buy products that create untaxed externalities like pollution or signaling.

Similarly, if you decide to relax and have fun, society can benefit (e.g. if you are participating in a community that benefits from having more members, or hanging out with your kids who enjoy your company, or doing unpaid work to improve the world) or suffer (e.g. if you contribute to congestion at a public park, or you drive around throwing eggs at people’s houses).

Overall I think that working is probably better for the world than leisure. The effect seems pretty small though (maybe 10% of the value you earn), and I think this isn’t a big deal compared to other efficiency considerations about a UBI. For example, it seems like it is smaller than any one of the additional overhead in additional redistributive programs, the costs of bullshit for benefit recipients, the efficiency losses inherent in poverty (e.g. from credit market failures), and the deadweight losses from taxation.

(Of course the calculus is different for people who pay taxes, since that’s pure social benefit, but a UBI should mostly change employment for low-income families who are paying very low tax rates.)

Dependency trap

Another concern is that people who don’t work won’t develop marketable skills, so they will remain trapped as dependents. Some of my thoughts on that concern:

  • On paper, it seems more likely to me that being poor would be a trap than that that having money would be a trap (e.g. by making it very difficult to invest for the future and forcing short-sighted decisions).
  • I haven’t seen evidence that giving people money significantly reduces their future earning potential and have seen weak evidence against.
  • I think the prior should be against paternalism; at a minimum we ought to have solid evidence that people are making a mistake before being willing to pay to incentivize them to do something for their own benefit.
  • If people decide not to work in the future because they expect to continue having a UBI, that’s not a trap, it’s just the temporally-extended version of the previous section.

Other effects

Funding a UBI would involve significant tax increases, and those really do inefficiently decrease taxpayer’s incentives to work. For example, paying $8k per adult in the US would require increase the average tax rate by ~15%. But quantitatively the disincentive seems moderate, and this isn’t relevant if we are comparing a UBI to other government spending. When people get really concerned about incentives to work it’s because they are thinking about beneficiaries who no longer have to work, and that’s the part I think is mistaken.

This argument is especially clear when we compare UBI to programs that add a bunch of machinery with the goal of incentivizing work. For example, families with a positive incentive from the EITC probably have an inefficiently large incentive to work since the size of the EITC seems to dominate plausible estimates of externalities (and then other families have an absurdly reduced incentive to work).

There may be strong economic cases for these policies based on empirical analyses of e.g. the benefits to communities from more people working. But I haven’t seen studies that convincingly assess causality, and it feels to me like public support for these policies is mostly based on unjustified pro-work sentiment.

One thought on “It’s not economically inefficient for a UBI to reduce recipient’s employment

  1. I don’t think economists make the error you criticize, but I’m not surprised that politicians and other decision-makers are motivated by the sense that work is virtuous and necessary. Many don’t understand the efficiency considerations, and those who do aren’t much motivated by them.

    Indeed the Finnish experiment’s research question was determined by parliament.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s