On Monday, I debated Professor Alan Manning of the London School of Economics on whether the federal minimum wage should be increased from $7.25 an hour to $15 an hour. Guess which side I took.
The audience was MBA students at Northwestern University’s Kellogg School of Management. The moderator was a Kellogg associate professor named Amanda Starc, whose area is health economics. She did a great job and, as I pointed out at the end, was more prepared as a moderator than I had seen in a long time. She even had her own slides as a basis for questioning both Alan and me. The discussion was quite civil.
I asked them to record the debate and they did so; I think it will be available soon for people to watch.
Alan Manning’s major claim was that even a substantial increase in the minimum wage will cause little or no job loss. In his argument, he gave the following hypothetical example. A low-skilled worker is producing something that nets the employer $12 an hour. Of course the employer isn’t going to pay $12 an hour because at that wage the employer would be indifferent between hiring or not hiring the worker.
So far, so good.
But then Professor Manning suggested that the employer would pay $8 an hour. So if the government then raises the minimum wage to $10 an hour, the employer will continue hiring the worker.
See the missing step? I did. So later, when I had a chance, I pointed it out. If the worker’s productivity is really worth $12 an hour, one would expect another employer to come along and offer $9. Then someone would offer $10. Then $11. I don’t know how close the wage would get to $12. But it’s highly implausible to think that it would stick at $8. If no one offered more than even $10, which already is $2 higher than in Alan’s hypothetical, then a potential employer is leaving $4,000 on the table annually.
Professor Starc suggested that I was proposing that the labor market is perfectly competitive. I responded that I wasn’t; I was making the much less ambitious claim that the labor market for relatively unskilled workers is competitive.
READER COMMENTS
Jon Murphy
May 12 2021 at 2:51pm
I think we see this trend in the data:
1) There’s a lot of voluntary turnover in the low-skilled occupations. This indicates to me that workers are searching for the highest wages. If they feel they are underpaid, they do elsewhere.
2) Workers who start out at minimum wage do not stay there very long.
john hare
May 12 2021 at 4:35pm
That assumes that low skilled workers are gaining skills that are worth more money. In my experience, many low skill workers don’t gain those skills. The ones that do become semi-skilled or skilled over a period of time. Go to almost any small business and ask if they can relate the term “30 year rookie”?
It really annoys me the people that stay on the bottom due largely to their own actions, and then blame others. Not to say that opportunity and ability is always available to some. There are a lot of road blocks and toll plazas on the road. Some will overcome and some won’t.
Off topic question. Lumber and steel prices have risen enormously in the last several months, by a factor of 3-4 in some cases. Is anyone using this as a foundation to explain and predict what will happen to those prices and availability over the next several months as compared to the shortages last year when prices were capped?
David Seltzer
May 13 2021 at 6:51pm
Jon, Let me ask an econ 101 question. Wouldn’t an employer pay $12 dollars and hour to produce the same in marginal revenue? We learned best profit comes when marginal costs just equal marginal revenue and capture the $4k.
Jon Murphy
May 14 2021 at 9:37am
Ultimately, yes. The firm would need to lure inputs (of which labor is one) from other uses, so they would need to offer a higher rate than the next-best alternative for the worker. Ultimately, that would equate to the worker’s marginal productivity (which is the firm’s marginal revenue).
Mark Brady
May 12 2021 at 5:17pm
David, you need to place a question mark in the title of your post!
More seriously, I suggest readers check out Alan Manning’s CV to get an idea of where he’s coming from.
https://personal.lse.ac.uk/manning/AlanManningCV.pdf
I look forward to listening to the debate.
David Henderson
May 12 2021 at 6:30pm
There IS a question mark there.
Mark Brady
May 13 2021 at 2:03pm
Now there is!
Daniel Kuehn
May 12 2021 at 6:31pm
Can you say a little about the difference you see between competitive and perfectly competitive. If you mean it’s imperfectly competitive then isn’t that all Manning is saying?
if you two are agreeing it’s not perfectly competitive but instead is imperfectly competitive then it just boils down to what numerical examples are empirically sensible right?
Bill
May 12 2021 at 8:03pm
I believe the issue was whether the labor market as a whole is perfectly competitive vs whether the labor market for relatively unskilled workers is competitive, i.e., is there monopsony in the market for unskilled workers.
Jon Murphy
May 13 2021 at 10:14am
Remember: competition is a process, not a state of affairs.
Daniel Kuehn
May 14 2021 at 9:05pm
John, correct but the process of competition is operating in both imperfect and perfect competition models so that doesn’t really clarify what I’m getting at.
Jon Murphy
May 15 2021 at 9:02am
Except it’s not. That’s part of Hayek’s point in that essay. There is no competition in the perfect or imperfect models (indeed, competitive behavior is often called anti-competitive behavior!)
Mark Barbieri
May 12 2021 at 6:46pm
“Alan Manning’s major claim was that even a substantial increase in the minimum wage will cause little or no job loss.”
I’ve never understood the stinginess of people that hold this belief and want a $15 minimum wage. Why not $100 or $500? I’ve yet to see any logic behind why $15 was chosen.
And what is the logic behind a national minimum wage? Does anyone seriously think we should have the same minimum wage for people in Silicon Valley or Manhattan and people in Puerto Rico or a rural town in South Texas? I would like a compromise. You can set your minimum wage at whatever level you think is appropriate for your needs, skills, location, and experience. If that’s $15/hr, great. If it’s $20/hr, even better for you. I just ask that you let everyone have the same right to set their own minimum wage.
Matthias
May 14 2021 at 12:00am
As far as I can tell, minimum wage laws are about enforcing a cartel with price discipline.
People having their own minimum reserve price wouldn’t keep others from breaking ranks.
(However, a federal minimum wage is still nuts. Even if you believe that a minimum wage is a good idea, that only justifies municipal, county or at most state wide minimum wage laws.)
Phil H
May 12 2021 at 10:53pm
I don’t even understand how this competition mechanism is supposed to work.
DH says:
“another employer to come along and offer $9. Then someone would offer $10…a potential employer is leaving $4,000 on the table annually.”
He seems to think that by paying higher wages, an employer would make more money. This is not correct.
Mike Hammock
May 13 2021 at 9:16am
It is correct if not paying higher wages means losing workers (or losing the best workers) to another firm (or worse, a competing firm). Your argument proves too much; if paying higher wages were never profitable, all employers would pay minimum wage.
David Henderson
May 13 2021 at 12:00pm
Mike,
Good point. I made that point in the debate.
Phil H
May 12 2021 at 11:14pm
I worked out the answer to my previous comment – it’s just a typo, right? You meant that the *employees* were leaving $4000 on the table.
But why? Why would anyone offer the employees $10 per hour, if they are able to pay just $8 dollars an hour? Our thought experiment starts with the employees being paid $8 per hour, so we know that it’s possible to hire at that rate. What is the motivation of the employers to pay more?
This hypothetical doesn’t work in the way DH thinks it does, and it’s to do with the replaceability of zero-skill labor (I’m assuming zero skills as a proxy for low skills). The point is that zero-skill labor is perfectly replaceable; and there’s always someone desperate for a job; so companies in general have the opportunity to force wages down. Another important point about zero-skill labor: the productivity of zero-skill labor doesn’t depend on the skills of the worker (we’ve already specified, they’re zero). Productivity (the amount of value an employer can get out of the employee per hour) depends entirely on the employer and their organizational efficiency.
So the correct way to understand minimum wage hikes is not about the employees at all. Rather, it is a government push to eliminate low-efficiency businesses that are unable to make effective use of the human capital in this country.
Jon Murphy
May 12 2021 at 11:25pm
But that’s the problem. Your hypothetical is an edge case. A zero-skill worker would indeed command a salary of $0 since he is perfectly replaceable. But low-skilled workers are not perfectly replaceable. Recall the hypothetical Manning uses:
Thus, the employee has skills that produce on the margin $12 per hour. That may be low-skilled, but it is not zero.
If the worker is underpaid (eg, only earning $8), then another employer can make profit by hiring them away at a higher wage: their marginal revenue increases by $12/hr. Likewise, the old firm’s marginal revenue decreases by $12/hr.*
So, what we actually see is you describing the very process that David Henderson is describing: wages move to the marginal productivity of labor. If the marginal productivity of labor is zero (as in your case), then the wage will move to zero. Using the exact same mechanisms, if the marginal productivity is higher, then wages also rise.
*NB we needn’t rely on the employer to do the hunting here. The employee may as well, if they correctly estimate their value. There are two sides to every transaction. That’s my point in my comment above.
Phil H
May 13 2021 at 1:54am
“A zero-skill worker would indeed command a salary of $0”
That’s not how skills work.
“$0 since he is perfectly replaceable.”
That’s not how anything works.
Jon Murphy
May 13 2021 at 8:58am
Yes it is. If you have nothing of value, people will pay you nothing.
It seems to me you’re being imprecise with your understanding of “skills.” Precisely define a skill
robc
May 13 2021 at 9:08am
Even the lowest of low skill jobs require 3 skills to succeed.
Show up on time. Work while on the clock. Don’t whine.
Honestly, for most entry level positions, if you do any 2 of the 3 you will be in good shape.
Jon Murphy
May 13 2021 at 10:13am
Exactly my point. And, as one shows up on time and works, one acquires more skills. If nothing else, one learns about the business they are working in. They acquire certain local knowledge that increases their skills and makes them more valuable to the employer.*
On a technical note, this is why I argue there is some monopoly power and monopsony power in the labor market. As people acquire local knowledge (both employees about their company and companies about their employees), it makes them unique and thus not “totally irreplaceable” for either party.
Jon Murphy
May 12 2021 at 11:28pm
Of course, the minimum wage by design and results do not make effective use of human capital (or machine capital). I mean, that’s just econ 101.
Henri Hein
May 13 2021 at 2:15am
No, the employer. Assume 2000 work hours per year. If there is an employee working at $8/hour producing $12/hour, then somebody else could hire them at $10/hour and make ($12-$10)*2000 per year, which works out to $4000.
I think the larger point is, given the large spread we are talking about, $7<<$15, it does not seem plausible that between thousands of employers, none of them are going after these profit opportunities.
Phil H
May 13 2021 at 2:22am
Right, I wondered if someone would go for this old fallacy.
“then somebody else could hire them at $10/hour”
Why would they have to? They could just hire someone else at $10 an hour. You’re smuggling in an assumption that the supply of labour is terribly limited. That assumption is unwarranted. There are millions of low-skilled unemployed people in America. The supply of people is not the limiting factor.
robc
May 13 2021 at 6:57am
Low skill employers are begging people to come to work. They aren’t available, pushing prices up. DQ in Outer Banks, NC was advertising $13/hr starting when I was there 2 weekends ago.
The competition that you say doesn’t exist clearly exists. Henderson is obviously right.
I have no idea who is making minimum wage right now. It would have to be in a very low cost-of-living area. Which goes back to the point made somewhere above about a national minimum wage being a dumb idea.
Jon Murphy
May 13 2021 at 10:10am
Phil-
To robc’s point, your explanation cannot explain what is really happening in the world. If your story is right, how do you contend with the fact that the vast majority of low-skilled workers (approx. 95% according to the BLS) are earning above minimum wage? NB: this is adjusting for variations in state minimum wage.
MarkW
May 13 2021 at 7:13am
Ah, the old Marxist ‘reserve army of the unemployed’ fallacy. As robc says, the defacto prevailing minimum wage is above $7.25 virtually everywhere except the poorest, lowest cost-of-living areas. In most of the U.S. the prevailing minimum for low-skilled labor is above the statutory minimum and has been for years. How do you think that happens if — as you apparently believe — there’s an unlimited supply of desperate prospective low-skilled employees who will take whatever pittance an employer offers?
Jon Murphy
May 13 2021 at 9:00am
Not terribly limited. Just limited. All you need for price formation is scarcity. Relative pricing will tell us how “terribly limited” the good is. And such scarcity does not solely depend on the absolute number of people. Remember Econ 101: for every transaction, there are two sides.
Phil H
May 13 2021 at 12:15pm
Er… what?
So, none of you are following the logic of the thought experiment DH started with. His initial starting assumption was that there were some people working for $8 per hour. I accepted that (and I think all of you above are neglecting that assumption).
His second assumption, unstated was that if another company wanted to pay some unskilled people $10 per hour, it would have to lure them away from the first company. I reject that assumption.
Now, to some of the odd assertions you make:
Robc: “Low skill employers are begging people to come to work.”
They should try paying more! This is a very curious situation. This website is supposed to be a bastion of people who believe in markets. But as soon as poor people make a market choice to not work, you all get very upset about it.
“The competition that you say doesn’t exist clearly exists.”
I didn’t say competition doesn’t exist. I just said, if you’re going to hire more imaginary workers, they aren’t necessarily going to be the same ones who were working at the first imaginary company.
Jon: “your explanation cannot explain what is really happening in the world”
I’m not sure what you think “my explanation” is. But I know that most of it is just your own imagination! Because I don’t think I’ve written anything about what I really think, yet. In this post I responded quite tightly to an error in what DH wrote.
MarkW: “Ah, the old Marxist ‘reserve army of the unemployed’ fallacy…unlimited supply of desperate prospective low-skilled employees who will take whatever pittance an employer offers?”
OK, now we’re getting to it. This is the interesting comment, and it’s where I think the problem lies with the argument.
Because we’ve all over-simplified. I understand why DH oversimplified, for the sake of a tractable thought experiment. But on this question, it ultimately just sets us thinking wrong. So here’s a bit of what I actually think:
In general, the wages required by low-skilled workers fall on some distribution. I imagine it as a bell curve, though I have no idea if the distribution looks like that in reality. Thus a very small number would be willing to work for $3/hr, most fall in at about $8-9/hr, and a few hold out for $12. Obviously, companies are interested in catching those willing to work for the least; companies face the problem that they can’t really pay different rates to different low-skilled people, because they’ll protest. So they need to set their rates at the minimum possible level so that they’re always able to hire as many as they need below that rate. So, size of company makes a difference, as does level of staff turnover that they’re willing to accept.
In general, I don’t see companies as competing directly with each other for workers. There is very rarely an absolute shortage of people. Instead, they compete indirectly, through the price mechanism; mainly, what companies are doing is competing with the distribution of wage expectations.
So, where I think DH went wrong is to assume that the market for low-skilled labour has cleared to such an extent that *all low-skilled people* demand the same wage. I don’t think they do; they fall on a distribution (wider or narrower) always. Thus, the existence of the company paying $10 per hour doesn’t mean that the company paying $8 cannot get workers. And we know this is true, because we know that different companies pay different wages. Obviously, even within the low-skill sector, there is variation in the wages and conditions that people will accept.
A minimum wage, then cuts off the wage acceptance distribution. It says, even those who would be willing to work at rates below X will not do so. I can understand why libertarians disagree with that. But a lot of the arguments introduced against minimum wages are pretty ropey – including the one in the OP here. Ultimately, I just see this as an empirical question: does the economy work better with a minimum wage or without one? And the evidence seems inconclusive. A $15 minimum wage, if introduced fast, would be a shock to rural America… I just don’t know if that’s a good or a bad thing.
robc
May 13 2021 at 1:33pm
Literally the very next sentence after the one you quoted says that is happening.
robc
May 13 2021 at 1:37pm
True, but some of them will be.
If company A is paying $8 and company B hires at $10, some of the people they hire will be people currently not working and some will be people moving from A to B.
No one is upset about it, that I have seen. I am upset about paying them to not work, but I am fine with them choosing to not work. I have done it myself in the past, for short periods of time. I didnt expect to receive anything for it (although I did collect 1 week of unemployment in 2019).
Jon Murphy
May 13 2021 at 1:43pm
What you said earlier:
This explanation cannot explain why employers pay higher wages. It cannot explain labor turnover.
robc
May 13 2021 at 1:46pm
That assumption is entirely within your head. Of course, there is a distribution. However, just because those only willing to work for $10 won’t work for $8, doesn’t mean that those willing to work for $8 wouldn’t prefer to work for $10.
If company A and B are otherwise the same, some of the $8 workers at A will apply for the $10 jobs at B. The math works out the exact same whether there is a distribution of minimally acceptable wages or they all demand same wage.
Jon Murphy
May 13 2021 at 2:01pm
I’m not sure where that assumption comes from given that Henderson (and Manning) explicitly say otherwise:
Again, we do not need perfectly competitive markets where everyone is a price-taker. All you need is competition, where everyone is a price-seeker.
Phil H
May 13 2021 at 2:35pm
Hi, rob.
“That assumption is entirely within your head. Of course, there is a distribution.”
No no no. There’s a lot of irrelevant stuff in this thread, but that point is the very nubbin.
DH’s hypothetical was a deliberate simplification, in order to explain his argument about why the minimum wage is a bad idea. There’s nothing wrong with arguing in that way. But in this particular case, this particular simplification is a problem (because, as you’ve agreed, in reality, the wages people demand fall on a distribution).
Here’s DH: “Professor Manning suggested that the employer would pay $8 an hour…another employer to come along and offer $9. Then someone would offer $10. Then $11. I don’t know how close the wage would get to $12. But it’s highly implausible to think that it would stick at $8.”
Note how he says “the wage”. In this simplified story, when one employer comes along and offers $9, all employees leave the $8 employers, and start working for $9. That becomes “the” new wage. This is very explicitly how DH’s simplified story works. If you don’t believe that the employees abandon the $8 company, then DH’s whole argument falls apart. (This is what he means at the end when he says that “the labor market for relatively unskilled workers is competitive”).
So this is what I’m disagreeing with. This particular construction of the argument. It may well be that you can de-simplify this argument, and run it using more realistic wage distributions. But DH didn’t do that here. He presented this simplified version of the argument, and it’s not a good one.
robc
May 13 2021 at 3:50pm
And they would. That isn’t the argument. No one would stay behind at $8 (assuming the 2nd company had enough spots) if they could get $9, assuming the jobs are the same (which isn’t a real situation, but not the one you are arguing against. Commute times would be different, company culture is different, etc — but, the simplification is that that doesnt matter). The distribution doesn’t mean that some people would prefer to work $8 instead of $9. It means that there might be a pool of people ONLY willing to work for $9. They would rather be unemployed than work for $8.
But if identical company B set up outside the company A parking lot and offered every employee a raise to do the exact same job, there isn’t any employee of A that would go “Nah, I prefer to work for less money.”
There is nothing in the “minimal amount to be willing to work” distribution that counters anything in David’s example.
Phil H
May 13 2021 at 4:22pm
Well, there’s all those real factors that you just ruled out. Some might not want to change jobs; they might want to keep working with their friends; or they may prefer that particular type of work. Or there may not be enough jobs at the second company… Those are exactly the factors that do exist in the real world, and they are the reason why DH’s story doesn’t work. The fact of a company offering jobs at $9 does not mean that no one will accept employment at $8. And we know this is true, because it happens in the real world.
DH’s story was meant to prove that if low-skilled employees *can* make $10, then in fact, they all *will* make $10 bucks (and he would go on to argue that the existence of $8 employees means therefore that they *cannot* make $10 bucks, hence the minimum wage is a bad idea for those people).
But that’s exactly what I’m saying is wrong. Because of all those real-world factors you mentioned, there are plenty of low-skilled people who *could* make $10; but in fact they only make $8. Because those $10-paying companies don’t usually set up in the car parks of their old employers.
Jon Murphy
May 14 2021 at 10:38am
Phil-
You fundamentally misunderstand the point of the reasoning; you have too dichotomous a way of thinking. It is incorrect to say:
Economics is about trends. Higher-skilled workers will tend to earn higher pay than lower-skilled workers. An employee who produces $12 will tend to earn a salary of $12. At any given point in time, could it be the case s/he is underpaid? Yes. But there is a profit opportunity there, and as long as both firms and individuals want to improve their lot, they will seek out the opportunities.
For your story to work, there must be only one single employer and literally no other options other than death for the employee. Then, and only then, can your “reserve army of the unemployed” story work.
The fundamental problem you’ve not acknowledged, never mind tried to answer, is that your story cannot explain the real world. It cannot explain why wages for the overwhelming majority of workers (even in low-wage jobs) are above minimum wage. It cannot explain the relatively high voluntary quit rates among the low-skilled. It cannot explain the lengths that firms go to in order to keep workers.
Thomas Lee Hutcheson
May 13 2021 at 8:12am
Too bad raising the incomes of low wage workers continues to be debated as between minimum wage levels instead of between minimum wage and a wage subsidy.
David Henderson
May 13 2021 at 12:02pm
Actually, that issue did come up. In response to a question, I said that I don’t want even an EITC but that an EITC would be much better than a $15 minimum. Wait til you see what Alan Manning said in criticizing an EITC.
David Henderson
May 13 2021 at 12:01pm
Thanks Jon Murphy, Mark W, robc, and especially Henri Hein for your cogent remarks.
Ahmed Fares
May 14 2021 at 11:29pm
I’ve read and argued it in the other direction.
A “greedy capitalist” hires a worker for $8 an hour and sells their output for $12 an hour and makes $4 an hour in the process. A “less-greedy capitalist” hires a worker from the reserve army of the unemployed, pays that worker $8 an hour and sells the output for $11 an hour, making $3 an hour off the worker and capturing market share away from the other capitalist. This competition between capitalists continues until all the excess profit is competed away.
Note that as prices fall, the real wage of the workers is rising.
robc
May 18 2021 at 11:21am
That fails the “real world” test, as my Dairy Queen example mentioned upthread shows. If DQ is offering $13 an hour, then the change is going the other direction, they are raising wages to pull people out of non-work or from competitors.
Your scenario is theoretically possible and could be happening at the same time (it is what is happening when a company moves a factory to Vietnam or Honduras — they are getting less productive workers but are paying less, theoretically for a net gain).
chris
May 16 2021 at 10:38am
I always find it amusing when libs argue the job market is not competitive (or not competitive enough) to ensure that low-skilled workers get paid what they’re owed.
If this were a real concern, wouldn’t you favor policies that make the market MORE competitive, not less?
Likewise, why would you assume that the government can and would set the right price (even if you assume the market has set the wrong price)? Surely, to use our given hypothetical, the government might set the price at $15 an hour, rather than $12. That is, there’s just no reason to think the market’s price is MORE erroneous than the government’s. Indeed, since the government’s price is likely to be a one-price-for-all, the government’s propensity for error is almost certainly greater.
Second, why would the government even attempt to set the right price? The government would be looking to set the politically-optimal price, not the economically-optimal one. Since there’s no necessary relationship between those two prices, it’d be sheer coincidence if the government sets the right price. It’s like somebody aiming at X; what’s the chance they’ll hit Y? Not so great unless you have a known, fixed relationship between X & Y, but here there is none – the politically-optimal price bears little or no known relationship to the economically-optimal price.
Long and short, the idea doesn’t relate to the policy proposal. Using minimum wage to help the poor is like using chemo to treat COVID. I can’t prove that chemo wouldn’t help somebody somewhere beat COVID, but it’d be blind luck if such irresponsible &
dangerous treatment actually worked.
Comments are closed.