Sunday, June 08, 2008

Downward Sloping Demand for Unskilled Workers: Employers WILL Respond to Minimum Wage Hikes

Angry Bear writes: We have a long record to compare the teenage unemployment rate and the minimum wage (see graph above). If you look at the two series you see a very inconsistent record. Sometimes a rise in the minimum wage is followed by a drop in the teenage unemployment rate and sometimes it is followed by a rise in the teenage unemployment rate. Essentially, the correlation between the teenage unemployment rate and changes in the minimum wage is zero, strongly implying that there is no causal relationship.

MP: Let's assume there is no correlation between: a) teenage unemployment rate and b) changes in the minimum wage. That is not the same thing as saying that "the minimum wage has no negative effect on teenage employment." Here's why:

Even if the same number of teenage workers are employed after a hike in the minimum wage, reflected in NO change in the teen jobless rate (and this is not necessarily true), there are many other adjustments that employers would make to offset the monetary increase in labor costs:

1. Fewer hours - unskilled workers might still be employed, but at a reduced number of hours (the BLS counts workers as "employed" even if they work 1 hour per week). Full-time workers now become part-time workers. Overtime hours are eliminated. Full-time workers now are forced to work a split-shift (e.g. 11 a.m. - 2 p.m. and 5 p.m. - 8 p.m.). Therefore, we would expect a negative relationship between increases in the minimum wage and HOURS WORKED.

2. Reduced benefits - employers can adjust "total compensation" and offset higher monetary wages by: a) no longer providing free uniforms, forcing employees to now pay for uniforms, b) no longer providing free or reduced food at fast food restaurants, c) reducing or eliminating "employee discounts" on the employer's merchandise, d) eliminating paid holidays, e) eliminating scholarship programs, f) eliminating group discounts available through large companies like McDonald's, g) eliminating employer sponsored or subsidized health care benefits, etc.

3. Fewer opportunities for advancement, fewer or reduced (or no) wage increases, fewer or reduced (or no) bonuses.

For example, see the list of benefits here for McDonald's workers in Canada (I couldn't find a comparable list for the U.S., but I assume it would be pretty similar), and you'll see that are at least ten non-monetary benefits offered to even unskilled minimum wage workers that could be adjusted in the face of higher monetary wage costs resulting from legislated minimum wage increases .

Bottom Line: Demand curves slope downward, and the market for unskilled workers is no exception. Employers WILL respond to increases in the minimum wage, in many ways that will NOT show up in the teenage unemployment rate, but still to the DISADVANTAGE of unskilled workers.


At 6/08/2008 10:23 AM, Blogger spencer said...

You are attacking a straw man and you know it. All your arguments are pure theory that have been countered by modern main stream research.

At 6/08/2008 10:35 AM, Blogger spencer said...

The only reason I did the post was to counter your post where you blamed the minimum wage for all of a change that was clearly dominated by normal business cycle developments.

In your post you said " For workers 25 years and over, the jobless rate has remained pretty stable"

In the month before the minimum wage took effect the adult unemployment rate was 3.5%. If you want to claim that an increase from 3.5% to 4.1% -- or about a one-third increase -- is stable that is fine by me But do not expect me to take your argument seriously. When I saw your chart it was obvious that you were playing with the scale to distort the change. You are very smart, and you know how to play the game of distortions that fool people that want to be fooled. But it does not work with everyone.

At 6/08/2008 10:37 AM, Anonymous Anonymous said...

You should at least admit that the lack of a correlation is significant evidence against the idea that minimum wage increases cause significant increases in employment. If a strong correlation were discovered, wouldn't you tout it as evidence for the opposite view? You can't have it both ways.

At 6/08/2008 12:27 PM, Anonymous Anonymous said...

So let's see, Spencer. The demand curve for labor slopes upward?

At 6/08/2008 1:29 PM, Anonymous Anonymous said...

Are you going to back up your assertions, Spencer?

If not, then they are pretty easy to ignore.

At 6/08/2008 3:28 PM, Anonymous Anonymous said...

"You are attacking a straw man and you know it."

This said by a man stuffed with straw.

At 6/08/2008 3:53 PM, Anonymous Anonymous said...

Spencer: 0.6%/3.5% equals about a sixth, but hey, if your incivility rests upon your use of numbers that are off by a factor of two, then your debate in valid only amongst liberals, socialists, and other anti-intellectual curiosities.

At 6/08/2008 4:30 PM, Anonymous Anonymous said...

"Essentially, the correlation between the teenage unemployment rate and changes in the minimum wage is zero, strongly implying that there is no causal relationship."

This is a totally irrelevant statement and MP is correct.

The correleation between those two series is not enough to reveal probable cause. You need to overlay other variables, such as GDP growth.

When economy is expanding, increases in minimum wages may not affact teenage unemploynment but that is not the case if there is a prospect of recession or slowing demand, rising inventories, etc.

Actually, if you take out the good times, the chart depicts exactly what MP is asserting.

Someone must learn about hidden variables. MP is correct.

At 6/08/2008 10:34 PM, Blogger bobble said...

it doesn't seem like a hike in the minimum wage can possibly have much effect on teenage unemployment, as not very many teens are paid minimum wage.

according to the BLS, in 2007: "Among employed teenagers paid by the hour, about 7 percent earned the minimum wage or less . . . "

in fact, the whole minimum wage argument seems irrelevant since the percentage of the hourly workforce paid the minimum wage has declined steadily from 13.4% in 1979 to only 2.3% in 2007


At 6/08/2008 11:34 PM, Blogger David A. Harding said...

If we reason deductively, we can quite easily prove that price floors (which includes the minimum wage), will always reduce demand (fewer employee working hours) -- with one exception: if the commodity (employees) were undervalued prior to the price floor and are undervalued or correctly valued after the price floor, then demand will stay constant.

In English (and somewhat inductively): I worked at McDonald's when I was 16, and I might have been worth more than the crummy $5/hour they paid me -- but, being 16, I was unwilling to negotiate for a higher wage. If most of the minimal wage workers at McDonald's were like me, working for less than we were worth, a small increase in the minimum wage could benefit us without reducing demand for our services.

The inductive evidence, for example: Card & Krueger, seems to back up the above story: small increases in the minimum wage benefit workers who don't negotiate well for their wages.


At 6/09/2008 8:24 AM, Blogger OBloodyHell said...

> You are attacking a straw man and you know it. All your arguments are pure theory that have been countered by modern main stream research.

You are not impressing me, spencer.

Even if "he knows it", others reading may not.

If you are going to make a claim, you need to actually support it by using at least a link or two to carry your own claim.

> In the month before the minimum wage took effect the adult unemployment rate was 3.5%. If you want to claim that an increase from 3.5% to 4.1% -- or about a one-third increase

Do you just make this stuff up in your head, spencer? You conveniently leave out whatever data source you produce your numbers from. Since you constantly do this I am left to wonder if you just make them up out of whole cloth.

From Table a-7, BLS data (non-seasonally adjusted unemployment, in K):
16-19: 1358/1082 == 25.5% increase from April to May.
20-24: 1584/1345 == 17.8% increase
25+: . 5470/5179 == 5.6% increase
All: . 8412/7606 == 10.6% increase

So the under 25 data almost doubles the increase in unemployment from April to May, all by itself. Not sure where Dr. Perry's numbers come from -- but you can see how mine were derived. Unless I'm misreading something in that BLS report, it's a fact that the unemployment of teens is a significant part of the current unemployment situation.

And it also seems fairly reasonable to bet that employers, knowing the increase was coming this month, stopped hiring new teens LAST month in anticipation of it.

(Note: Since some states already have higher wage laws, such as Washington, Mass, and Cali, the effect of the new wage laws should be insignificant in those states, and employment should be otherwise unaffected there. THIS would be a good test of the notion. It should be possible to show how those places (presumably already adjusted to the higher wage) are not as affected by this increase. One would expect a lower increase in teen unemployment for May & June.

There also have got to be studies which show the effects of such higher minimum wages compared to adjacent and non-adjacent states. There are certainly obvious effects one would predict if teen employment rates go down with increased wage floors. So it should be easy to cite some studies which show this to be true. One would also expect some visible behaviors for border regions, where teens can cross a state boundary to find work in a lower minumum wage.

> This said by a man stuffed with straw.

Oh, he's stuffed with something, but I'd suggest that it's hardly straw...

At 6/09/2008 5:24 PM, Blogger juandos said...

Hey spencer, did you know that YOUR socialistic world view and economics seem to be at odds with each other?

Get yourself a dose of Walter Williams for a bit of the real world sir...

At 6/09/2008 5:34 PM, Blogger Marko said...

I think the unemployment of teens probably is also influenced by the release of the latest installment of the Grand Theft Auto video game.

Don't tell me you really think that increases in the minimum wage don't affect employment. Really.

Oh, and "working for less than you are worth" is impossible. The true value of a commodity can be perfectly determined by the price at which it sold. You can call that Marko's Axiom if you like.

At 6/09/2008 9:10 PM, Blogger David A. Harding said...


Marko's Axiom requires an infinitely precise monetary unit and probably a perfectly competitive market. For example, let's look at the effect of inflation on wages. Every day in the U.S., inflation reduces the value of current prices slightly, but most employers only increase the price of wages once a year. Apply Marko's Axiom and you'd determine that workers decrease in value 364 days a the year and increase in value one day a year. Or you would determine that inflation only happens once a year. Either way, no economist today will take you seriously.



Post a Comment

<< Home