Minimum Wage, Shminimum Shmage!

The fight for fifteen! What a fantastic slogan! Let’s look at the policies behind it.

There’s a lot of fuss over the results of the economic research regarding minimum wage, and I’m not going to link it because you have Google; you can do that yourself. Basically every study finds more or less the same thing.

The general facts are (and several of them should be pretty obvious):

1. In general, employment effects of minimum wage increases have been negative, but very small.
2. The employment effects are bigger for bigger changes.
3. The employment effects are bigger for changes closer to the median wage in the area.
4. The employment effects are highly concentrated in a few populations, particularly teenagers, but also recent immigrants and ex-cons.

This is kind of a mystery. Maybe it isn’t a mystery to you, but let it sit with you for a second.

Okay, did you let it sit?

What might have popped into your head is this: Suppose you work at the minimum wage: $9 an hour. Now, the minimum wage rises to $10. You go look for a new job, and they hire you for the new minimum wage: $10.

So why wouldn’t they hire you before?

You would think, naively, that if someone is willing to pay you $10 today, then surely someone else would have been willing to pay $9.75 even before the minimum wage was raised, and you would have quit your old job working for $9 and made $.75 more an hour. 

But that didn’t happen. As we discussed earlier, you were working at the minimum wage before (keep up).

So, one of two things must be true:

1. There were none of the well paying jobs before the minimum wage was raised, but now there are.
2. The well-paying job was always waiting there for me, but, like the boy in the John Hughes movie, I only noticed how beautiful she was after some inciting incident.

Neither of these are very satisfying explanations. The second one is mildly insulting. Perhaps the first makes more sense. But… where did the job come from? Here’s an explanation: They didn’t have to pay well before, so they didn’t. Now they have to, so, fuck it.

But this explanation, like bargain cosmetics, is less alluring on deeper investigation than at first view. There are lots of companies that hire people. They can’t just pay what they want. Otherwise, everyone would be paid the minimum wage. If there were an activity that could be productive paying people $10 an hour, and it only paid $9 an hour, then more companies would hire people to do that activity and the wage would be bid up until there were no more opportunities worth hiring people for.

So, if there is a job that only pays $9 an hour, then a lot of the jobs there are only worth paying someone $9 to do. If you had to pay someone $10 to do them, it wouldn’t be worth it, and those people would be fired. The fired people would then be unable to find jobs elsewhere.

Except… No. That just… Doesn’t happen.

It should! It should! And it does happen a little bit. Just not very much. As I said before, the employment effects of raises in the minimum wage tend to be very small.

Somehow, when the minimum wage rises, there are suddenly just as many good new jobs as bad old jobs.

There are a few explanations that float around for this:

1. The increase in labor demand from the rising minimum wage (and thus more money in workers’ pockets) offsets the slide along the labor demand curve that comes from increasing wages.
2. Employers have monopsony power over wages. So, they hold down wages, but when you legally bar them, they can’t anymore.

The first explanation I am not going to get into, because demand side changes do not fucking matter outside of recessions. Nothing more needs to be said about that.

The second explanation is more plausible (And treated in much more depth in the literature). But it should jar you, because the idea of a single small restaurant in a big city being monopsonistic is weird.

How does a monopsonist hold down wages? By restricting demand. You would think, if this were true, that companies would actively work to restrict the supply of jobs, so they can pay low wages, so they can hire for low wages. This can happen in a few instances, such as when one large company in a small town replaces several smaller companies – a situation in which wages tend to fall. 

But could that really happen in Seattle? In big, dense industries like restaurants and health care that have hundreds if not thousands of firms? Unlikely. Companies in a big city are not actively stopping themselves from making profitable hires so that they can keep down wages. They’re too little; they would just be pointlessly shrinking their operations.

What’s more likely is that a single employer can hold down the wage of a single employee by not quite paying them their worth and praying the employee doesn’t bother to go out and get a better job. This is… Probably what happens. A lot of the (small) increases in unemployment that you see after minimum wage hikes comes from increases in search time: how long a person looks for a job before they find one.

That makes sense. Suppose you’re looking for a job. A job comes along. You stop searching for jobs. That job might be a good one or a bad one. You don’t really know. So, if the government instead made all the bad jobs illegal, you would have to search for longer, but all the jobs around would be good ones. Basically, people were being underpaid before because they didn’t know how much they were worth, or the hassle of finding a new job made them unable/unwilling to get paid their worth. This definitely helps explain why minimum wages don’t reduce unemployment.

But there are some other phenomena that can’t explain:

1. Raises in the minimum wage tend to move employment from one sector to another.
2. Raises in the minimum wage tend to decrease employment at certain types of employers (like small businesses) and offset that by increasing it at others.

You can square those with a story about monopsony power (be my guest; give it a try!) but it’s hard. So, it’s safe to presume that another force keeps labor demand inelastic: some jobs pay more because they suck, and you can do those jobs if the old ones disappear.

Remember the possibility from way earlier that there were better paying jobs all along, I just didn’t take them? We dismissed that possibility, because it would be dumb not to take that job, so obviously I would have taken it.

Well, it wouldn’t be dumb not to take that better paying job if that job is shitty.

There are some jobs you can take to increase your salary, even if you have a very low education: logging, oil rigging, guarding prisons. These jobs are easy to get and pay very well for the education required. They also tend to be the jobs that people shift into when minimum wages rise (well, employment in childcare and food service falls, which is more or less the same thing). So why didn’t people do these jobs until the worse paying ones disappeared?

Because they suck. I, for one, do not want to be a prison guard. I work in an office. Sure, I complain about my boss, but, fundamentally, when I walk in at 9 am, no one is going to stab me.

Obviously this is an extreme example, but others are more typical. Moving from a family business to a big-box store, you’re monitored more carefully and often have irregular, algorithmically determined hours. It sucks. But the pay’s usually better.

Minimum wages don’t have to force low-productivity workers out of the market because those workers can become high productivity by going to jobs that suck more.

So, at bottom, this is why minimum wages don’t reduce unemployment:

1. People don’t know what they’re worth, so get paid less than they should.
2. Almost anybody can do other jobs that are less pleasant but more productive if their current job disappears.

But this gets us to a problem.

You see, minimum wage research is kind of like drunk guys looking for their wallets under the stoplight because everywhere else is too dark to see.

Fundamentally, you can only research increases in the minimum wage in places where there are increases in the minimum wage. These are places like: Massachusetts, New York, New Jersey, California, Washington, and Chicago. You can’t study minimum wage increases in Alabama. The Alabama State Senate gets in the way. 

You should be noticing a pattern here. The pattern is that these are all weird labor markets, and unfortunately, they’re labor markets where both of the following tend to be true more often than in the rest of the country.

1. People don’t know what they’re worth, so get paid less than they should.
2. Almost anybody can do other jobs that are less pleasant but more productive if their current job disappears.

Why? Let’s tackle the first. These are markets with high and rising costs of living, where few jobs are below $15, and where the poor and less educated are leaving en masse for the sun belt. That means that nominal wages can rise for poor people while real wages hold steady or fall, due to continuous gradual increases in rent. In a situation like that, all factors that use rent as an input should rise in price (including labor), otherwise real wages would fall. 

Now, we’re good Keynesians, so we know that labor responds to nominal wages, not real wages. Partly because of contract setting, partly because of strikes, partly because psychology, labor bargains for nominal wages, not real wages. This means that their real wages can get out of whack from the value workers are actually providing. Basically, when real wages are changing in a different direction from nominal wages people don’t know their worth. What was a good salary one year might not be one five years later, or even two years later if a city gentrifies fast enough. So we shouldn’t be that surprised if labor markets in a place like New York aren’t paying people for their full productivity.

But that doesn’t apply as much in a normal place. In a normal place (Like, say, North Carolina) people have a much better idea of what wages mean and what wages they can get. That’s why they moved to North Carolina.

So the other thing: in a place with a high cost of living, almost no-one making $8 an hour is doing a very unpleasant job. Very few people in places like Boston make that little. The only ones who do are both nearly unemployable and doing something that’s not too awful (like working as a Home Health aide). Almost no one in those places who makes $10 an hour is really maximizing their income. They have other priorities.

But in lower-wage, lower-cost places, there are lots of people who are working awful jobs and still only making <$13 dollars an hour: poultry processors, Amazon warehouse workers, etc. These tend to be the same workers who are forced out of work by minimum wage hikes in high-cost places: ex-cons, immigrants, and teenagers. These are people doing everything to be as productive as possible and still  they make less than $15 an hour.

So we have our two reasons that big minimum wage hikes haven’t cause unemployment in the studies we’ve done:

1. People in the kinds of places that have minimum wage hikes often don’t know what they’re worth, so get paid less than they should.
2. Almost anybody in the kinds of places that have minimum wage hikes can do other jobs that are less pleasant but more productive if their current job disappears.

But in a lot of the country, the parts we haven’t studied, neither of those are true.

[Edit: Here’s something else. In places where land is expensive, anything that makes business more expensive should almost entirely translate to a reduction in rents (AKA the price of land) not a decrease in employment. Why? If you increase cost of business (say, by raising the minimum wage), you reduce profits. But land is still expensive, meaning other businesses still want to be there. Since profits have fallen, fewer businesses want to be there, but in the end, the same number of businesses will be operating, because there’s only so much land. So, rents must fall to accommodate. Hypothetically, businesses could move to have fewer employees per acre, but that’s difficult. So, again, we expect minimum wages to have no impacts on employment in the place we can study (high cost places), while they might have large impacts in the places we can’t (low cost places).]