Paul Krugman suggests that cutting the minimum wage won’t help employment. My hope here is to uncover his errors.
First, Krugman makes the argument for cutting the minimum wage. He states:
Here’s how the fallacy [of composition] works: if some subset of the work force accepts lower wages, it can gain jobs. If workers in the widget industry take a pay cut, this will lead to lower prices of widgets relative to other things, so people will buy more widgets, hence more employment.
Of course, that’s not the argument that is typically made at all.Most students of Principles of Microeconomics should have heard the argument against the minimum wage. The real story goes like this:
Demand for labor is downward sloping. The supply of labor is upward sloping. This establishes an equilibrium wage where the quantity demanded and supplied are equal. A minimum wage, if it does anything at all, pushes the wage above this level, so that there is a greater quantity supplied than demanded. We call this surplus supply “unemployment”.
The argument doesn’t rely whatsoever on changes in the relative prices of products. Rather, it relies on the fact that the demand for labor is downward sloping – which comes from marginal productivity, and that the supply of labor is upward sloping – which comes from, for example, “reservation wages” being met for more workers.
Back in the day, Keynes suggested that this story is inadequate at the macro level. The problem is that the marginal revenue product of labor depends on demand for the product – but that demand depends on income, and income depends on wages. So, if wages fall, then incomes fall, which drives down the demand for the product – which, in turn, drives down the demand for labor – in the same proportion that wages fell. So, in the end, there’s no impact on employment. The decrease in wages caused a corresponding decrease in employment.
However, within the Keynesian framework there is a way for cutting wages to still “fix unemployment”. The fall in prices means that the real value of money has increased – so people have excess real money, and try to divest themselves of this extra money by buying bonds. Buying bonds pushes down the interest rate, which, in turn, stimulates investment, which, in turn, stimulates employment.
Krugman says that this story doesn’t work once we’re in a liquidity trap. Interest rates are already at their lower bound, so buying more bonds won’t decrease them – and therefore, investment doesn’t get stimulated, and therefore employment doesn’t increase.
So, where, specifically, does Krugman go wrong?
(1) The demand for labor depends on the expected discounted marginal revenue product. So, what matters is not the present demand for the product, but the expected future demand. For Keynes’s original argument to work, we need a feedback from present prices to expected future prices, so that as one falls the other does so – and does so proportionally. This is not necessarily true – in fact, it’s probably not true.
Suppose that right now the price of oil is $50 a barrel, and I expect it to go up to $100 a barrel next year. Then, tomorrow, the price of oil shoots up to $75 a barrel. Do I revise my expected future price up to $150, to keep the proportions the same? Maybe, maybe not. But, I will say that “not” certainly feels more likely.
So, what happens if demand doesn’t fall in proportion with the current price of the product? Employment increases as wages fall. Even if a decreased current price arising from the fall in income does decrease the expected price to some degree, as long as it is less than the fall in present wages, the level of employment will increase.
(2) The entire apparatus fails to realize that there is significant variation in wages across workers – even within the same industry or even the same firm.
Why does this matter? Because it breaks the argument down entirely. For example, suppose that workers are divided into groups based on seniority. In that case, the fact that new workers are hired at lower wages than before doesn’t mean that ALL workers are paid lower wages. So, wages don’t fall in proportion with the minimum wage – which means that incomes don’t fall by that proportion, so demand for the product doesn’t fall in proportion, so product prices don’t fall in proportion, so demand for labor doesn’t fall in proportion. In the end, what does this mean? In short, that the Principles of Microeconomics story is basically right. When marginal wages fall, employment goes up – and the fact that not all wages move together prevents the Keynesian feedback.
(3) The framework ignores that excess real balances can be drawn down by buying things.
To be fair, Krugman acknowledges this, but writes it off as insignificant. Here’s what he says:
Somebody is going to ask, what about the real balance effect? Doesn’t a falling price level make people wealthy, by raising the real value of the money they hold. The answer is, consider the magnitudes. Before the crisis, the monetary base — the system’s “outside money” — was around $800 billion. (It’s a much more confusing situation now, so I won’t try to parse the current numbers here). This means that even a 10 percent fall in the price level, which is very hard to achieve, would raise real wealth by only $80 billion. Compare this with the effects of the decline in housing and stock prices, which reduced household wealth by $13 trillion in 2008. The real balance effect is totally trivial.
The problem? Krugman looks at the wrong number. It’s not the monetary base (reserves + currency) that matters. It’s the quantity of money balances, something more like M1, MZM, M2, or the Austrian Money Supply. Let’s just look at MZM (which includes checking accounts, savings accounts, and money market accounts). Right now, MZM sits at about $9.5 trillion. So, a decrease in prices of 10% would lead to an increase in wealth of about $950 billion. True, this is less than 10% of the $13 trillion decline, but it’s not “totally trivial”.
Why does this matter? Because it’s one more reason that a decrease in wages won’t lead to a proportional decrease in prices – which means that a decrease in wages will increase employment.
A significant part of the Keynesian argument against free markets (and for minimum wages) rests on the argument that Krugman presents. The assumptions that this argument rests on are simply not true. Meanwhile, the assumptions underlying the argument for lower wages increasing employment are far more reasonable.
{ 29 comments }
Lucas, your theoretical argument seems sound, and I applaud you for it.
However, it does not address our current situation. Our persistently high unemployment is the result of many factors, the minimum wage being a fairly small part. Here are other, more important (IMO) factors:
1) Unemployment Insurance
2) Low Income Subsidies (food stamps, etc)
3) Government borrowing crowding out private investment
4) Frictional unemployment from malinvestment corrections
5) Regime Uncertainty
6) The creation of high-wage government jobs
7) Anticipation of taxes on business
Particularly where a business can outsource labor positions to a foreign nation without such interventionism, expect unemployment.
Can you cite the source for the 7 other reasons for unemployment
…I just saw that the Krugman article is specifically focused on the minimum wage…
So my prior comment likely sounds stupid now.
In any case, I think Mises.org would do well to explain to the layman where the unemployment in this crisis came from.
This entire refutation of Krugman’s nonsense is to long and unnecessary. All that is needed to be told is that:
Cutting the minimum wage is NOT the same thing as wages falling. You put more people to productive work, who before were basically producing nothing and were even welfare consumers, and you get more net productivity, and not less. Total demand increases as a result of increase in productivity.
Krugman doesn’t even understand the “problem” he’s trying to explain.
Of course the minimum wages law causes economic distortions. But once again its a distortion that results from other bigger economic distortions like interest rate targeting and money printing by Congress and the Federal Reserve which goes to special interests. The minimum wage laws, rent control laws, unions asking for higher wages, people refusing to work for some wages, etc is all related to the decline in the value of money. The value of our money today is controlled by the Treasury, Federal Reserve and Commercial banks. So if money and credit creation leads to higher prices for everything with wages of course always lagging nowadays, then people who could earn the same or more on unemployment will stay on unemployment for as long as they can.
The marginal productivity of labor depends on one additional factor: a corresponding increase in capital. (Labor is useless without capital they can use.)
This means that if capital investment is impossible, the demand for labor is not downward sloping but limited at the current capital stock. So, supposing fast food chains could hire employees at any wage they want, they could not clear the market for labor unless they could open more stores, which requires investment, which requires savings, which is exactly what there is no longer any of in a recession.
Krugman is a puppet on JMK’s knee where he learned his mentor’s rhetorical tricks. Keynes *refuted* Say’s law by defining it in a way that would make Say blanch. Keynes was not the first to employ straw men in his “economic” disputes, but he was the Master of its use.
And let’s not forget the basic difference between Keynesian and Austrian economics: that something must be produced before it can be consumed.
stranger says:
“(Labor is useless without capital they can use.)”
back-massages, baby-sitting, domestic work…? given its non-specific nature, seems to me there’s almost an endless demand for labour at the right price, even with a given fixed capital stock (isolating one business doesn’t really enlighten us about the economy).
instead of me doing the lawn, i pay someone else to use the same capital (my mower). my shopping is done by someone else using my capital (car). my house can be vacuumed not by me, but by someone else, subject to price. at any given time, there’s a vast amount of unused capital which can be brought into active service when labour prices make it economic to do so.
DD,
Thank you for simple language and simple common sense, lost arts, it seems.
“(2) The entire apparatus fails to realize that there is significant variation in wages across workers – even within the same industry or even the same firm.”
How can there be people doing the same work yet getting different pay rates? ‘Seniority’ doesn’t exist in a free market – it’s union-imposed entity. If an employer found he can get workers who will do the same work for $10/hr instead of the current $20/hr then he drops the wages and/or hires the $10/hr workers. News story have shown people are currently willing to take a pay cut over unemployment ($10/hr is better than $0/hr).
I’m having a harder and harder time understanding the labor market. Labor, for the most part, is seen as homogeneous, and claims like “fast food chains could hire employees at any wage they want” are heard frequently. But our labor force is extremely specialized and faces rigidities other than trade union participation and minimum wage. If credit induced booms create a demand for a specialized job, say, finance, the labor force will respond. And when the boom comes to end, laborers from the finance sector simply aren’t going to flood the fast-food market. They spent too much money and 4 years in college learning about financial markets and instruments. I think Austrians should investigate how changes in the structure of production also effect the labor market–creating additional rigidities. If they haven’t already.
This is the last straw. Never before have I seen a writer so blatantly use ambiguities in language in an attempt to deceive:
“…if some subset of the work force accepts lower wages, it can gain jobs.”
There is absolutely no reason to refer to a “subset of the work force” collectively other than to obfuscate the point that it’s precisely those people who AREN’T working that would be able to “gain jobs.” He equivocates between the working work force in the first part of the sentence and the non-working work force in the second. Are people really dumb enough to fall for this semantic manipulation?
“Are people really dumb enough to fall for this semantic manipulation?” Yes. I mean, it’s worked pretty well for Krugman and the Keynesians up to now, innit?
This post has been linked for the HOT5 Daily 12/17/2009, at The Unreligious Right
Lucas,
You lost me at the outset with this statement.
“Demand for labor is downward sloping. The supply of labor is upward sloping. This establishes an equilibrium wage where the quantity demanded and supplied are equal.”
In pictures or in words, a curve by itself has no meaning in economics. It must still be explained in other terms. With the other terms, why do you need the curve? If demand goes down as price goes up, why not just say so, why that “the demand curve slopes downward?†Why translate from straightforward language into a circular maze, and meaning into non-meaning?
Confusion is complicated, economics simple, and, if it isn’t simple, it isn’t economics. And this discussion certainly hasn’t been simple. Let me try to make it so.
The market tends toward equilibrium between the supply of and demand for labor, as for anything else, and, interference with it, toward disequilibrium, which, in the case of a minimum wage law, would be an oversupply of labor and unemployment.
As DD pointed out, eliminating the minimum wage law simply eliminates the unemployment. There is not less but more production and more overall demand.
This was the core of the Keynesian argument, as you related it:
“…if wages fall, then incomes fall, which drives down the demand for the product – which, in turn, drives down the demand for labor.â€
Wages per se fall only when production per se falls. If production per se is constant, wages per se are constant. If some are falling, others are rising.
Even if all wages were falling, in money terms, because, say, people started saving more money in mattresses, there would be no less purchasing power, demand, and employment. Prices and wages would fall. The purchasing power of each dollar would rise. There would be just as much purchasing power, demand, and employment as before, just at lower prices and wages.
Gil,
I can only speak for my profession (software quality assurance), but in that field people doing the same work definitely make different salaries.
I get a pretty good wage despite not having a college degree, because I have 25 years experience and some useful certifications (e.g. from the American Society for Quality). I show up on time (something of a dying art) and I mentor less experienced workers.
Other people bring their own advantages such as experience with automated tools. Bottom line: we have several people doing the same job, but some do it better than others, and the pay scales (to some degree) reflect that.
DG Lesvic,
“In pictures or in words, a curve by itself has no meaning in economics.”
I definitely agree. My purpose here was just to present the “typical” Principles-level argument. Experience suggests that my statement is very close to what Principles-level classes teach. Of course, one can (and should) invest the curve with meaning.
The difficulty with your argument about tendency toward equilibrium eliminating unemployment is that Krugman, as a Keynesian, would say that there is such a thing as “equilibrium involuntary unemployment” that can’t be solved by lower wages – taking the idea straight from Keynes. The argument runs more or less along the lines that I presented. Simply claiming that tending toward equilibrium eliminates unemployment in the face of that argument could (I think rightly) be seen as simply refusing to deal with the argument that was presented. The argument presented here was largely an attempt to deal with Krugman “on his own grounds”, simply taking the standard Keynesian argument and tweaking it a few places to show its fragility.
Now, I do have to disagree with your statement re: production and wages. Consider a very simple example of Robinson Crusoe. He fishes for one hour, and that gets him 5 fish. His real wage then is 5 fish per hour. He fishes a second hour. Assuming that diminishing marginal returns is true (which is reasonable), he catches fewer than 5 fish – say 3. So, if he works 1 hour, production is 5 fish, and his real wage is 5 fish per hour. If he works 2 hours, production is 8 fish, but his real wage is 4 fish per hour. So, production and wages can move in opposite directions. Though I feel like we might have a difference of definitions…
Gil,
The problem: workers aren’t homogeneous.
Actually, I think it’s perfectly possible that pay could increase by seniority on a free market for a reason like this: Experience generally increases productivity. More “senior” workers have more experience and therefore are generally more productive than new workers. If it’s expensive to determine a worker’s level of individual productivity, it may be that the best option is just to approximate productivity with seniority.
Put differently, the different workers aren’t actually doing the “same” work.
I know it says “post an intelligent comment” but…um
duh why don’t we make minimum wage $1m a minute and make everyone rich and fix the economy? duh…..
simple
where’s MY nobel prize in economics?
Funny thing is, Dave, I’ve never seen a useful answer to that question. About the minimum wage, not your Nobel
Sandwich : I know, it’s crazy. For every argument against removing minimum wage, you simply take the opposite tack and see if it makes andy sense….from this article for example:
“The problem is that the marginal revenue product of labor depends on demand for the product – but that demand depends on income, and income depends on wages. So, if wages RISE, then incomes RISES, which drives UP the demand for the product – which, in turn, drives UP the demand for labor – in the same proportion that wages ROSE. So, in the end, there’s A POSITIVE impact on employment. The INCREASE in wages caused a corresponding INCREASE in employment.”
It’s saying the same thing…so raise the minimum wage to infinity and we will have 0% unemployment forever (trust me).
Newsweek – if you’re reading this, give me a call…I’m available for the cover shoot this weekend.
And I wonder if Krugman would take the same logic on something like…hmmm…I dunno….health insurance…
Have a health insurance premium floor of let’s call it $10m per family per month.
This should have no effect or a positive effect on health insurance coverage.
Why? Because if health insurance companies are earning so much per family policy, they can pay their workers and shareholders more who then demand other products in the economy, bidding them up and everyone else’s wages ad infinitum and eventually we are all rich with universal coverage.
Lucas,
Thank you for your very thoughtful response.
You wrote,
“Of course, one can (and should) invest the curve with meaning.”
No. One should dispense with it altogether. This is economics, not geometry.
You said that I failed to address the Keynesian point. I thought I had done so. I’d be most grateful if you could tell me exactly wherein I failed to do so.
As I pointed out, even saving money in a mattress does not withdraw purchasing power from circulation. With fewer dollars in circulation, prices will fall, and the purchasing power of the dollars remaining in circulation will rise correspondingly.
And so long as labor is scarce relative to land, the owners of land must still employ all of the available labor in order to maximize the profitable use of their land. The prices or wages at which they do so is immaterial.
What more need be said?
I don’t know what your fish story has to do with anything. Granted that at some point it would pay to switch from fishing to some other productive activity, or from activity altogether to leisure.
How does that justify a minimum wage law?
Lucas,
Thank you for your very thoughtful response.
You wrote,
“Of course, one can (and should) invest the curve with meaning.”
No. One should dispense with it altogether. This is economics, not geometry.
You said that I failed to address the Keynesian point. I thought I had done so. I’d be most grateful if you could tell me exactly wherein I failed to do so.
As I pointed out, even saving money in a mattress does not withdraw purchasing power from circulation. With fewer dollars in circulation, prices will fall, and the purchasing power of the dollars remaining in circulation will rise correspondingly.
And so long as labor is scarce relative to land, the owners of land must still employ all of the available labor in order to maximize the profitable use of their land. The prices or wages at which they do so is immaterial.
What more need be said?
I don’t know what your fish story has to do with anything. Granted that at some point it would pay to switch from fishing to some other productive activity, or from activity altogether to leisure.
How does that justify a minimum wage law?
Lucas,
Thank you for your very thoughtful response.
You wrote,
“Of course, one can (and should) invest the curve with meaning.”
No. One should dispense with it altogether. This is economics, not geometry.
You said that I failed to address the Keynesian point. I thought I had done so. I’d be most grateful if you could tell me exactly wherein I failed to do so.
As I pointed out, even saving money in a mattress does not withdraw purchasing power from circulation. With fewer dollars in circulation, prices will fall, and the purchasing power of the dollars remaining in circulation will rise correspondingly.
And so long as labor is scarce relative to land, the owners of land must still employ all of the available labor in order to maximize the profitable use of their land. The prices or wages at which they do so is immaterial.
What more need be said?
I don’t know what your fish story has to do with anything. Granted that at some point it would pay to switch from fishing to some other productive activity, or from activity altogether to leisure.
How does that justify a minimum wage law?
I completely disagree, minimum wage only hurts small business looking to take on employees either part time or for temp help. Right now jobs are almost impossible to find so if this was removed more jobs would be available, what is better, having a job or no job and not being able to feed your family? aaron kocourek
Remarkable blog you have here! greatly enjoyed the blog post I am ing to.
Thanks for the interesting information that ipicked up!
Comments on this entry are closed.