The minimum wage

Will Wilkinson offers a good summary of economic thought on the topic.  His early salvo:

The law of demand is very, very empirically adequate as it is…

Here is my previous post on the minimum wage.  Here is Tyrone on the minimum wage.

Addendum: Jane Galt (where is Jennifer?) adds commentary.


What is the "law of demand" that he is talking about? I don't think I've ever seen it written down and I have taken the odd economics class in my time. In any case, "demand for labour" isn't really demand because in general people don't consume labour. The amount of labour employed is determined by a production function, not a demand function.


The "turn down the AC" argument is just a way to illustrate adjustments on other margins. I'd illustrate it differently by instead illustrating it by an employer who increases the speed of an assemby line.

Incidentally, I believe that Richard McKenzie should be given credit for presenting this argument to the profession in his often overlooked 1980 article: Journal of Labor Research, v1, n2 (Fall): 255-64. I believe it is the best theory article available on the minimum wage. The last time I checked the Citation index, however, it had only been cited once since 1980; the cite was not by Card and Kreuger obviously.

I had no idea that the Card & Kreuger study was a telephone survey of fast-food joints. What a load of crap.

- Josh

The case for the minimum wage being a benign way to help those who would otherwise fall below it, and not just another way for those above it to shut out competition, seems very parallel to the case for protectionism driven by heartfelt concern for the plight of foreign low-wage competition. (The theoretical justification seems substantially the same, and the population of chests in which the feeling hearts beat seem to be substantially the same too.)

As a nonexpert, reading popular accounts, I notice what seems to be a peculiar revealed-preference or revealed-belief pattern in the protectionism case. Unions enthusiastically lobby for protectionism that would cut off their competition, out of their sincere belief that by so doing they help those poor foreigners improve their lives. But I have never heard of US union representatives going to Japan or Germany to testify in favor of restricting imports from unionized industries in the US (auto, steel) until the US increases its interventions in its labor markets in ways that Japan and Germany intervene and that US unions admire. So I wonder, do I seem to see this pattern only because I read the wrong news sources, and too few of them at that? (I doubt it, wouldn't such visits be controversial enough to report?) Or is it a real pattern, and US unions really do avoid helping themselves in this way, and help others only? (But I have to doubt that too; no one criticizes the unions for what almost looks as though it might be insincerity in their concern for their low-wage competition?)

(I have been wondering this for some time, and hanging a question on an article from GMU about the minimum wage seems at least as likely to get me an answer as any other method I can think of, especially since Robin "sincerity of disagreements wrapped in mind-bending math" Hanson guest-blogs...)

Somehow, its argued that:

1. A Large Illegal Immigrant population does not lower wages,
2. but small increases in the minimum wage would decrease employment and cause great distress.

Its gotta be one or the other. It can't be both, unless Econ 101 isn't the final word on all of this. If Econ 101 isn't the final word, I want that same argument, what ever it is, to be applied equally in both cases.

We know that artifically high wages from a minimum wage law doens't effect total compensation all that much because so many people are paid minimum wage. It probably lowers the number of people working. It doesn't alter the total compensation all that much.

I can see an argument for relaxed immigration as it doenst really effect wages with a call for a higher minimum wage, as it wouldn't really effect employment. In fact, thats what I support. Unless the laws of supply and demand have been repealed, that is.


You wrote:
"if a rise in the minimum wage leads to a speeding up of
production line the result is a rise in productivity that justifies the
higher real wage the minimum wage employee is earning. That may be more a good thing then a bad thing."

Not so if we begin with compensating wage differentials in equilibrium. Some people prefer slow work rates and are willing to accept much lower compensation for such rates. For example, professors of economics work at a much more relaxed pace than an economist working for a drug company (I'm a professor and a guy I graduated with works for a drug company; I earn much less).

So the minimum wage may close off options that workers preferred, making them worse off. Workers who wanted to make wage concessions for more relaxed environments at work will be denied. These deals were made by firms, not because they liked having people work slowly, but because the wage concessions were more than enough to make it worth while to the companies.

If the minimum wage forecloses deals that were mutually beneficial to firms and workers, resources are less efficiently allocated to the detriment of society.

If workers are willing to take sub-min. wage wages to get training, why cut off such opportunities?

Etc., etc.

Mickslam, the immigration question is not as simple as you make it sound, because there are huge effects in both directions: the immigrants are not just suppliers of labor, but consumers of the things produced by labor, and producers of things which are in some cases complements of the things produced by other laborers, and there are other things like scale effects too.

Do you really think that Econ 101 predicts that if the Americas were uninhabited until Robinson Crusoe landed somewhere on them, his income (in any useful sense, like availability of all the goods he needed to live, not in some silly sense like "I claimed title to 20 more virgin square miles today! woohoo, I'm *so* rich!") would be enormously greater than if he landed in the Americas as they are today? Unless you believe that, then you should realize that you have not expressed yourself very clearly, because your argument makes it sound as though you believe it.

The minimum wage question is slightly complicated by all sorts of little things, but it's hard to find huge compensating effects. Just as pushing your immigration analysis to the Robinson Crusoe extreme illustrates that your analysis is ignoring some huge effects running the opposite direction, pushing the minimum wage to an extreme like a billion uninflated dollars per hour illustrates that the minimum wage critics' analysis is consistent with the huge effects: push it hard enough, and anyone can see that that the effects the critics worry about must dominate. (Or would you like to argue that unemployment would not be disastrously increased by such an extreme minimum wage, so much that the unemployment effect dwarfed all other considerations?)


Are you sure that Card's immigration stuff is being embraced any
more than his other stuff?


You wrote:

"Robert Solow, put it: 'The main thing about this research is that the evidence of job loss is weak. And the fact that the evidence is weak suggests that the impact on jobs is small.'"

This is exactly what Richard McKenzie's theory explains : Journal of Labor Research, v1, n2 (Fall 1980): 255-64. I believe it is the best theory article available on the minimum wage. If you had picked up on my previous post, you probably would not have written your last post.


You wrote:

"The usual suspects predicted massive job losses among those affected by the increase from $4.25 to the current level of $5.15. Instead, low-wage workers experienced the strongest job market in 30 years. Poverty fell to historic lows, particularly for the most disadvantaged workers, such as less-skilled minorities and single-mothers."

It appears that you are falling victim to the classical error of mistakenly seeing causation in things due to time sequence; the following is from Wikipedia:

"Post hoc ergo propter hoc is Latin for 'after this, therefore because of this.' It is often shortened to simply post hoc. Some philosophy books translate the Latin to simply: 'If after, then therefore, because.'
Post hoc, also known as 'coincidental correlation' or 'false cause,' is a logical fallacy which assumes or asserts that if one event happens after another, then the first must be the cause of the second. It is a particularly tempting error because temporal sequence is integral to causality — it is true that a cause always happens before its effect. The fallacy lies in coming to a conclusion based only on the order of events, which is not an accurate indicator. That is to say, it is not always true that the first event caused the second event."


sorry about the tone of my post - it reads more harshly than I intended.

jim -- is there data to support your argument that the existence of the minimum
wage prevents people from accepting sub-minimum wage jobs because of the
training that it provides. I know about the type of programs many skilled
labor unions like plummers run to train young people, but I have never heard of
the minimum wage having an impact on it.

What are some example of what you are talking about?


You wrote:

"is there data to support your argument that the existence of the minimum
wage prevents people from accepting sub-minimum wage jobs because of the
training that it provides."

It is difficult to get data on labor contracts that would have occurred, but
now don't because of the minumum wage. Still there are other types of evidence, supporting
the McKenzie thesis. For one, Unions support increases in the minimum wage, but as you
point out union workers already earn more than the minimum typically. So why do they
favor increasing it every chance they get? Because trained non-union labor is a fairly
good substitute for union labor whereas untrained non-union labor is not. Better substitutes
tend to make the elasticity of demand for anything more elastic. More elastic demand for union
labor means that the union's ability to drive wages up without cutting employment too much is

Alan Blinder published a survey of economist that found the average response
was a 10% rise in the minimum wage causes a 1% drop in employment.

this estimate seems to be in the range most of the people here are discussing.

this means that out of every 100 minimum wage employees 99 will get a 10% raise
and one will lose his job.

If my objective function is to maximize the welfare of minimum wage employees
I will take this trade off every time.

Orginally you have 100 people working a 40 out week at $5.00 an hour.
this generates income of $20,000.
After the wage hike you have 99 people working a 40 hour week at $5.50 an hour.
This generates an income of $21,780, a 8.9% increase.

If we use the old econ profs argument about how much the winners would have to
bribe the losers to accept this trade off it turns out if the winner each give 1%
of their new income -- so their effective raise is 7.9% rather then 8.9% --
the individual that loses his job is also better off.

the math is simple. If the percent increase in the minimum wage is larger
then the percent drop in employment the welfare of minimum wage employees
is improved.

Consequently, to prove your argument that a rise in the minimum wage makes minimum wage
employees worse off you must demonstrate that it generates a larger percent drop in
minimum wage employment then the percent increase in the minimum wage.

P.S. even the guys at CATO assume that the increase in the minimum wage will be larger then the drop in employment.

Comments for this post are closed