let’s suppose that in my town, eliminating the minimum wage laws would cause the equilibrium wage for entry level jobs to fall to $9/hour. (I think it would he higher, but I’ll use this example.) The government could then chip in enough wage subsidy to boost the equilibrium wage to say $14/hour, which is about $28,000/year for a full time worker.
1. Note that the wage subsidy has to go to more than just minimum-wage workers. Otherwise, the marginal tax rate on workers earning just above the minimum is too high. You can only phase out the subsidy slowly, or not at all. I am sure Scott was thinking that, but he didn’t say it.
2. But why go to the trouble of implementing a wage subsidy? We already have a wage penalty, namely the payroll tax. Just cut the darn thing. You only need to implement a subsidy if cutting the payroll tax to zero still doesn’t get the equilibrium wage up to your desired minimum.
So taxpayers would pay 100% of minimum wage unless employers choose to contribute something?
No. Unless there’s no phase-out, the supply-demand effects for low-wage workers still apply. Imagine Employer A offers $0/hr with the government picking up $14/hr. Then Employer B opens a new facility and offers $1.50/hr. With the gradual phase-out, this results in an effective, say, $15.25 to the employee. So Employer B is able to cherry-pick A’s best workers. But then Empoyer C comes along….
However, if the subsidy is too generous relative to the equilibrium wage, then there will be serious problems. Imagine that what Employer C does to compete with A and B above is to offer $0/hr….but with 26 weeks of paid vacation a year, thereby magically turning that $14/hr subsidy into $28/hr. If government is picking up too much of the wage, then some employers will start to compete by making jobs cushier rather than higher paying.
In the comments (at 1:30am), Sumner mentions that he meant for the wage subsidies to be on a sliding scale. That makes things a little more complicated, but also potentially addresses the marginal tax rate issue.
One thing that’s annoying about many anti-poverty proposals is that they tend to focus on seemingly less-controversial nominal wage matters without getting into explicit discussions of fundamental goals and assumptions about the labor characteristics of the intended beneficiaries. What are we really and specifically trying to do and why? Who are we trying to help? Why do they need help? How can we be sure this will really help them, and even if we are, that we won’t cause other problems as side-effects of the intervention?
Plenty of people have tried to get into these weeds, David Friedman among them, and I think a fair summary of their attempts to get advocates to be clear and specific and then also explain and justify those particular goals, is that this is one of those social phenomena that one is just supposed to be savvy enough to know that the words people use aren’t supposed to be questioned as to actual logic, evidence, analysis, and details, or taken too seriously and too literally. It’s a combination of “showing that you care” and an effective political formula in a democratic society.
Economics and policy wonks, however, tend to run with it anyway, and because numbers are the nails to their hammers, they quickly make it all about the numbers, ignoring or assuming away a lot of the human factors and frictions that are at the root of the status quo, which are obscured when one simply accepts that a market equilibrium is somehow at the ‘wrong’ place.
Arnold is very much correct — eliminate the payroll tax (& all total taxes?), on a sliding scale such that at $15/hr some percentage of the taxes is paid (100% or less), and at some lower level, maybe $10/hr, 0% of the payroll taxes are taken out of the pay and sent to the gov’t, instead they go to the employee. In order to have a stronger support, I’d set the top at $16/hr, and the bottome at $6/hr, with a 1% increase in the gov’t tax paid for each 10 cents ($0.10) of more pay. At $8/hr, they only pay 20%; at $12/hr they only pay 50%.
Scott Sumner’s fine reply was to the Utopian Scott Alexander’s excellent but long:
<a href="http://slatestarcodex.com/2018/05/16/basic-income-not-basic-jobs-against-hijacking-utopia" title ="Basic Income, Not Basic Jobs"
http://slatestarcodex.com/2018/05/16/basic-income-not-basic-jobs-against-hijacking-utopia
I’m certain the basic income won’t work now, so basic jobs / a job guarantee is better.
The Levy Institute should be mentioned here.
http://www.levyinstitute.org/pubs/op_55.pdf
(testing the ref tags which I stopped using a while ago)
(test 2, maybe following directions better)