Wednesday, December 4, 2013

Universal income versus minimum wage

The concept of universal basic income (a.k.a. guaranteed minimum income) got a bit of traction in the media when the Swiss proposed in October to put it to referendum. The idea is to give a check every month to every adult citizen, regardless of age, income, wealth, job status, or health condition. The intention is ostensibly to put a floor on the standard of living. My question in this blog post is: how would universal basic income compare with raising the minimum wage? The key to either policy, I think, is: what do we do with the existing tax and welfare system?
The likely first reaction to the basic-income scheme is to worry about disincentives to work. The likely second reaction is to mind the cost to taxpayers. If it comes down to a choice between universal income or our current patchwork of tax deductions and welfare schemes, universal income is clearly the winner. If I remember correctly from my undergrad econ, a lump sum tax or subsidy is not distortionary at the margin, whereas our ecosystem of taxes and welfare programs clearly is: it affects our decisions to work, save, and so on. Moreover, it costs a lot of money to administer all those welfare programs--clearly more than mailing a check to every adult every month. Universal minimum income would have an income effect, but if the level of the stipend is low enough it should not discourage most people from working. On the demand side of the labor market I do not see any meaningful effect.
To make universal income fiscally neutral (not to mention politically admissible) we would have to erase at least some of the current tax exemptions and deductions, and part of the welfare system. Suppose we started on a clean slate, we gave $10,000 a year to every adult, indexed for inflation, and we axed every existing welfare program (everything: disability, unemployment, food stamps, earned income tax credit, Medicaid, Medicare, subsidized housing, school vouchers, etc.). Tyler Cowen makes a very good point when he worries that we would quickly start amending such system, either on the grounds of social justice or public choice:
"Might there be circumstances when we would want to pay some individuals more than others?  Many critics for instance worry that a guaranteed income would excessively reduce the incentive to work.  So it might be proposed that the payment be somewhat higher if low income individuals go get a job.  That also will make the system more financially sustainable.  But wait — that’s the Earned Income Tax Credit, albeit with modifications. 
Might we also wish to pay more to some individuals with disabilities, perhaps say to help them afford expensive wheelchairs?  Maybe so.  But wait — that’s called disability insurance (modified, again) and it is run through the Social Security Administration.
As long as we are moving toward more cash transfers, why don’t we substitute cash transfers for some or all of Medicare and Medicaid health insurance coverage benefits, especially for lower-value ailments?  But then we are paying more cash to the sick individuals.  That doesn’t have to be a mistake, but it does mean that an initially simple, “dogmatic” payment scheme now has multiplied into a rather complex form of social welfare assistance, contingent on just about every relevant factor one might care to cite."
Adding to Tyler's list: Should we give a higher stipend to people living in New York than to those in New Orleans? The costs of living are quite different across locations. Should an adult with two children get a bigger payment than one with no dependents? Should we tack on a supplement when a natural disaster strikes? How about naturalized citizens? Should they start getting a check the day they get their U.S. passports, after a few years, or have a phase-in period? 
From a political standpoint it is clear to me that a pure universal income scheme is not feasible and that the only alternative is a modified program, in which case most of its benefits over the current system vanish. Political feasibility, on the other hand, is the main selling point of the minimum wage. We have lived with it since 1938, and it has been routinely raised with little controversy (at least compared to the Affordable Care Act or other novel schemes). The public opinion is that the minimum wage is a matter of social justice and that it does not cost the taxpayer a dime. Another contrast with the universal income idea is that it agrees with the hard-working-man ethic: raising the minimum wage might encourage people to work more. Convenient as these perceptions may be to politicians, they are partly wrong, because of the welfare and tax programs in place.
What happens to labor supply if we raise the minimum wage? Income taxes and the income and wealth thresholds for welfare programs start biting, so labor supply declines at the margin--and since minimum-wage jobs are hourly paid, many workers would have the flexibility to cut hours just below the point where they are still eligible for welfare. At very low levels of hours per day, workers would drop out of the labor force entirely and take up welfare payments.
On the demand side, economists often worry that a higher minimum wage will destroy jobs, or at least reduce the number of hours per worker. At least at the margin, I am not convinced that this is the main concern. Sixty-two percent of workers earning the minimum wage or less have service occupations (44% in food preparation and serving). Another 15% are in "sales and related" occupations, according to a 2012 report from the Bureau of Labor Statistics (table 4). By industry (table 5), 51% of minimum-wage earners are in the hospitality industry, and 16% in retail. It seems to me that replacing waiters, cooks, maids, and salespeople with technology is quite difficult (although not impossible: don't forget all those call center jobs outsourced to India, or the expanding use of self-checkout in supermarkets, or the introduction of robots to care for elderly people, for instance). 
I suppose that some employers would be forced out of business if we raised the minimum wage. However, if it were a binding constraint, we should observe "bunching up" around the minimum wage in the distribution of wages. What I see is the opposite: workers at or below the minimum wage have been overall declining since 1979, both as a proportion of all workers and as a proportion of hourly paid workers (about 3% and 5%, respectively, in 2012, table 10). That is quite remarkable, especially considering that in real terms the minimum wage has been generally declining. My view, therefore, is that minimum wage jobs have been disappearing because of changes in the industry and occupation composition of the labor market and that, for the majority of employers and positions, the minimum wage is not a binding constraint.
When considered against the background of the current welfare and tax systems, neither universal income nor the minimum is a clear winner. Raising the minimum wage is probably the path of least resistance, but it would not do much to put a floor on the standard of living, for a couple of reasons. First, the poorest of the poor are those completely cut off from the labor market, either because of disability, age, having a criminal record, etc. As unemployable, they do not benefit from a higher minimum wage. Second, the minimum wage is becoming a binding constraint for a declining share of jobs. Besides, when considering the rising cost of health care, a marginal raise in the minimum wage is hardly a ticket out of poverty--at least not at the current level of the minimum wage.
The best, politically-constrained option* might be a combination of: a streamlined welfare system (fewer programs, fewer conditions, lower income replacement ratios, and lower implicit tax rates); a universal minimum income, at a subsistence level; and a higher minimum wage. Even then, opportunities for loopholes, patches and exceptions in the universal income scheme are too abundant and I suspect that, within a few years, the system would degenerate into the jumble that we have now, plus the universal income program--which by then would have ardent defenders.
*Even better, increase investment in public goods, such as education and infrastructure, that will increase the long-term potential growth rate of the economy, and change drug and incarceration policies. That should go a long way towards reducing poverty. I'm not sure, however, that this is possible, even under loose political constraints.

1 comment:

Anonymous said...

Awesome article. Strange enough that there aren't more comparisons of minimum wage vs UBI out there.