One of the treats of the recent Liberty Fund colloquium on the Austrian and Chicago schools of thought was getting to read or reread various excerpts from Friedrich Hayek’s work. In a chapter titled “Government Policy and the Market” from his 1982 book Law, Legislation, and Liberty, Volume 3, Hayek nicely puts perfect competition in perspective and lays out the incredible benefits of real-world, as opposed to “perfect,” competition.
On the problem with perfect competition as a normative standard:
From basing the argument for the market on this special case of ‘perfect’ competition it is, however, not far to the realization that it is an exceptional case approached in only a few instances, and that, in consequence, if the case for competition rested on what it achieves under these special conditions, the case for it as a general principle would be very weak indeed. The setting of a wholly unrealistic, over-high standard of what competition should achieve thus often leads to an erroneously low estimate of what in fact it does achieve. (p. 66)
Hayek goes on to lay this out more.
Two comments:
First, I occasionally run into people who took economics as undergraduates, and even some how minored or majored in economics, who believe that because the market is not perfectly competitive, it has failed and that the door is wide open for government to intervene and improve things.
Second, one reason I like Shark Tank is that the sharks generally appreciate markets as they are. They will often ask those who pitch their firms and products what their margins are. The margin they have in mind often seems to be the difference between average variable cost and price and sometimes seems to be the difference between average total cost and price. If someone answered, for the latter case, that the margin is zero (which would be the case for perfect competition) all 5 sharks would say, almost in unison, “I’m out.”
On temporary monopoly as an incentive to innovate:
The inducement to improve the manner of production will often consist in the fact that whoever does so first will thereby gain a temporary profit. Many of the improvements of production are due to each striving for profits even though he knows that they will only be temporary and last only so long as he leads. (p. 70)
When I was teaching this point to my students and we were studying efficiency of perfect competition versus efficiency of real-world competition, I would ask them to imagine 2 buttons, one of which they could push. The first button would yield a world in which no one would make above-normal profits, even for a short time. The second button would yield a world in which such profits would be allowed. Which one would they push if they cared about long-run well-being? Most of them got that they should push the second button because pushing the first would reduce the incentive to innovate, thus reducing innovation, so that most innovations would come about randomly rather than due to focused effort.
The injustice of requiring a monopolist to produce to where the price equals marginal cost:
Quite apart from the practical difficulty of ascertaining whether such a de facto monopolist does extend his production to the point at which prices will only just cover marginal costs, it is by no means clear that to require him to do so could be reconciled with the general principles of just conduct on which the market order rests. So far as his monopoly is a result of his superior skill or of the possession of some factor of production uniquely suitable for the product in question, this would hardly be equitable. At least so long as we allow persons possessing special skills or unique objects not to use them at all, it would be paradoxical that as soon as they use them for commercial purposes, they should be required to use them to the greatest possible extent. We have no more justification for prescribing how intensively anyone must use his skill or his possessions than we have for prohibiting him from using his skill for solving crossword puzzles or his capital for acquiring a collection of postage stamps. (pp. 71-72)
Later, Hayek gives what he thinks is the clinching reductio ad absurdum:
There exists no more an argument in justice, or a moral case, against such a monopolist making a monopoly profit than there is against anyone who decides that he will work no more than he finds worth his while. (p. 72)
I would agree with Hayek that this is a slam-dunk reductio ad absurdum. But two authors have recently taken that absurd conclusion and run with it. In their 2018 book, Radical Markets: Uprooting Capitalism and Democracy for a Just Society, Eric A. Posner and E. Glen Weyl warm to a related proposal. Here’s what I wrote in my 2018 review of their book in Regulation:
Toward the end of the book, they even toy with having people pay taxes on their human capital. They give an example of a surgeon who announces that she would perform gallbladder surgery for $2,000 and pay a tax accordingly. She would be obligated to provide that surgery to anyone willing to pay $2,000. So if the surgeon was thinking of retiring, forget it. The only satisfactory solution for her would be to estimate the value of her services at a number that really would make her indifferent between working and retiring.
The authors are aware that they’re treading on sensitive ground here, writing, “A COST [common ownership self-assessed tax] on human capital might be perceived as a kind of slavery.” Might be? They claim that such a perception is incorrect, but the reasoning behind their claim is weak.
They implicitly admit that their proposal is coercive when they write that it would be a mistake “to think that the current system is not coercive.” How is the current system coercive? Here’s how: “Those with fewer marketable skills are given a stark choice: undergo harsh labor conditions for low pay, starve, or submit to the many indignities of life on welfare.” In short, to Posner and Weyl, being relatively poor is akin to being coerced. I would bet that a newly freed slave in 1865, though almost certainly poor, would understand the difference between poverty and coercion better than Posner and Weyl seem to.
I’ll have more to say on Hayek on these issues soon.
READER COMMENTS
Andrew_FL
Apr 26 2023 at 9:01am
It’s actually worse than that. If no one ever earns above normal profits even for a short period of time, it implies a total lack of entrepreneurial discovery. Either no new profit opportunities emerge in such a world, ever, or no one ever realizes them. That’s a static world of no change and no progress.
David Henderson
Apr 26 2023 at 11:21am
Actually, I said it the way I did on purpose. There’s not, as you say, “a total lack of entrepreneurial discovery” but, rather, much, much less. Why not a total lack? Because some entrepreneurs will discover new products because they find it fulfilling. Not many. But not zero either.
Andrew_FL
Apr 26 2023 at 1:37pm
I think I’m interpreting this differently than you intended. If new products can still be discovered, why, in the hypothetical, do they not result in any above normal profits? In the hypothetical, do they never come into existence (what I thought you meant) or are they taxed away (what I now think you meant)?
David Henderson
Apr 26 2023 at 2:15pm
Ah, now I see your point. I’m saying that they’re taxed away right away. I should have made that clear.
Knut P. Heen
Apr 26 2023 at 10:24am
The monopoly model rests heavily on a linear demand curve. Suppose instead the demand curve is a rectangular hyperbola of the type p = k/q. The monopoly’s revenue is independent of q, p*q = k. The entire MR=MC argument falls to the floor because MR = 0. The interesting point is that this is the only true monopoly (because if you lose revenue by increasing price, there must be competitors around).
Jon Murphy
Apr 26 2023 at 10:37am
I’ll admit that I am not the most mathy person, so it’s probable I have a fundamental misunderstanding, but I just don’t follow your argument here.
First, are you defining P as price and Q as quantity? What’s k here?
Second (and assuming P = price and q = quantity), it doesn’t make sense to me that revenue is independent of q given that revenue is defined as p*q.
Third, you write:
Not at all. Demand curves slope downward for many reasons and the presence of substitutes is just one of those reasons. A bigger one is diminishing marginal returns, not to mention budget constraints.
Mike Hammock
Apr 26 2023 at 3:56pm
Pedantic side note: You mean diminishing marginal utility, not diminishing returns. Diminishing returns refers to inevitable eventual reduction of marginal product when using more of a variable input in the presence of fixed inputs.
Jon Murphy
Apr 26 2023 at 4:04pm
Yes, thank you (though diminishing marginal returns can play in as well for a downward sloping demand curve for a firm).
Knut P. Heen
Apr 28 2023 at 9:09am
I should have added that k is a constant, p and q are price and quantity.
Revenue = p*q = k/q * q = k
You get this demand curve if the utility function of the consumer is a Cobb-Douglas function (k is then a function of the budget and the exponents of the utility function). You do not need “crazy” assumptions to get here.
I don’t understand your third point. Either the money must go to the monopoly or to something else. The something else is the competitor.
Jon Murphy
Apr 28 2023 at 10:20am
Thank you for clarifying. It’s been a bit since I’ve done Cobb-Douglass. Now that you reminded me, I see what you’re saying.
Or there are the presence of dominishing marginal utility or budget constraints.
Knut P. Heen
Apr 28 2023 at 11:09am
I am not disputing that budget and diminishing marginal utility may affect the demand curve. I am saying that a monopoly is competing with something if a higher price leads people to buy something else instead. You are not really a monopoly if you are competing with something else, are you? Suppose Pepsi raises its price. Does it really matter whether I buy Coca-Cola or a new computer instead? Competition is competition.
David Henderson
Apr 26 2023 at 11:23am
If, as you say, the demand curve is a rectangular hyperbola, then you’re right that MR equals zero throughout. But MC exceeds zero. Therefore output is zero. You won’t observe such monopolies because they won’t produce.
Knut P. Heen
Apr 28 2023 at 8:49am
Almost correct. It is a cost minimizing problem and you may need to use the boundary conditions to find profit maximum (zero output gives zero profit, one or more units may produce positive profits because the price is very high). Cobb-Douglas utility actually produces rectangular hyperbolas.
A more general point is that a demand curve shaped like a hyperbola has a flat interval and a steep interval. The producer must decide how to set up the business, focus on the price insensitive consumers (steep part) or focus on the price sensitive consumers (flat part). If you do the latter, you are essentially facing a horizontal demand-curve similar to perfect competition and you will behave as if you were in perfect competition even as a monopoly. It thus follows that you need both a monopoly and a particular shape of the demand curve to generate the deadweight loss people worry about.
David Henderson
Apr 28 2023 at 5:03pm
You’re right that I’m almost correct. The right answer is that Q = 1. The price that leads to that quantity demanded leads to the same revenue as every other quantity but 0. So if there are any marginal costs at all, the firm will price to sell 1 unit.
Roger McKinney
Apr 26 2023 at 10:33am
I’m in the healthcare insurance industry and can see attempts by the state to make it perfectly competitive. We are limited now to just two means of adjusting for risk, age and gender. We can’t consider anything else. For the ACA plans, we are given three levels of benefits, gold, silver and bronze and the state dictates what benefits must be in each. And our highest rate can’t be more than three times the lowest. This forces young people to subsidize older ones.
The goal appears to make all insurance plans as homogeneous as wheat and force the industry to compete only on price.
robc
Apr 26 2023 at 1:07pm
gender or sex?
Can a person reduce their rates by timing their “change” of genders? Choose to be male when young and then female after age 45 or so?
Roger McKinney
Apr 26 2023 at 1:58pm
Haha! That hasn’t come up yet. But in today’s environment I would guess so.
Jon Murphy
Apr 26 2023 at 10:49am
That misconception is because many Principles textbooks define market failure that way. For example, in Dirk Mateer and Lee Coppock’s textbook Principles of Microeconomics, they define market failure as “an inefficient allocation of resources in a market” (pg 220 of the 3rd Edition, COVID-19 Update). They define an efficient market as a perfectly competitive market (pg 325). Monopolies, monopnopistic competition, and anything that is violates the assumptions of a perfectly competitive market are treated as per se market failures.
David Seltzer
Apr 26 2023 at 5:07pm
Jon: Firms are incentivized by profits to innovate. It seems, to me at least, perfect competition is less efficient than imperfect competition in the long run.
Jon Murphy
Apr 26 2023 at 5:11pm
Agreed
Rebes
Apr 26 2023 at 11:25am
I am not an economist, but I teach a class on energy policy at a law school.
Whenever a student starts an argument or proposal with “in a perfect [or ideal] world,” I tell them to stop right there and rethink what they were about to say, because the perfect or ideal state of affairs never exists and policy design is all about dealing with imperfections. Which may result in no policy if the imperfect status quo is superior to an imperfect policy.
Walter Boggs
Apr 26 2023 at 2:18pm
Agreed. In a perfect world, we wouldn’t have to say how things would be in a perfect world.
nobody.really
Apr 26 2023 at 6:50pm
What’s the difference between theory and practice? In theory, there’s no difference. In practice, there is.
nobody.really
Apr 26 2023 at 7:17pm
This is not entirely hypothetical. Many states have prohibitions on “price gouging.”
Indeed, some have argued that the failure of Texas’s electrical grid during February 2021 was due to a market for wholesale electricity that limited how high prices could go during emergencies. As a result, firms lacked adequate incentive to make the necessary (but expensive) investments to ensure that their generators would be able to operate during extreme (but rare) weather events.
nobody.really
Apr 26 2023 at 7:23pm
Henry George argued for financing government through a land tax. But what if a landowner has no revenue—in part because the landowner wishes to leave her land pristine and unoccupied? I would expect the state to pursue legal claims seize any assets available. Would that constitute slavery?
If we can countenance taxing land, why not human capital?
In many states, people getting divorced must fill out worksheets that reflect how much they have been earning, and divorce decrees often direct one party to make continuous payments to another party based on an assumption that the parties’ incomes will continue. As far as I know, you have no obligation to maintain that level of income—or any level. But failure to make the prescribed support payments may result in fines, a finding of contempt of court, and jail time. Is this slavery?
And how far do we want to take this? Should the state punish parents who fail to care for their kids? Negligent parents are merely choosing how to allocate their own resources, including human resources. The idea that parents should have to expend resources on their kids is just fuzzy-headed sentimentality, right? I expect you disagree–but can you articulate a rationale?
vince
Apr 26 2023 at 9:00pm
If someone doesn’t want to pay the land tax, he can avoid the tax by selling it. If he doesn’t want to work, he doesn’t have that choice.
nobody.really
Apr 26 2023 at 10:53pm
Well, kinda. The person would have to live somewhere, and that person’s landlord would presumably bear the tax on his behalf. (Then again, would a sea stead pay a land tax? Someone on the space station?)
vince
Apr 26 2023 at 11:11pm
A unique trait of the land tax is that the tax can’t be passed from landlord to tenant. Say the land rents for $1000, the land value is ten times the rent, and the tax is 5 percent. If the landlord passes the $500 tax to the tenant, the rent is now $1500, the land value $15,000, and the tax $750. Passing the tax to the tenant is a loss to the landlord.
nobody.really
Apr 27 2023 at 1:21am
If you structure a land tax that way, then the original hypothetical poses no problem: A person could entirely evade the land tax by not using the land at all, thereby generating no rent. But in your formula, use does not seem relevant, only rent. Most homeowners do not charge themselves rent–yet I suspect most homeowners remain liable for property taxes.
Moreover, in the real world landlords pay property taxes yet somehow recoup their costs.
I surmise all these mysteries are resolved when we realize that real-world taxing authorities establish a lot’s “assessed evaluation” using factors that do not rely very heavily on each individual lot’s actual rental revenue.
The Massachusetts Institute of Technology summarized one of its student’s 2017 PhD dissertation as follows:
Finally, recall that when Mariah Carrie started spending holidays in Bedford Falls, NY, her friends tried to lure her to build a house there by giving her some undeveloped property–but she declined. Media reported that she was concerned about the property taxes, but she actually said that she simply did not want a lot for Christmas….
nobody.really
Apr 27 2023 at 1:26am
Uh–make that Mariah Carey and Bedford Corners. So much for my knowledge of pop culture….
robc
Apr 27 2023 at 9:25am
That is not how the land tax works, as the “economic rent” is independent of use.
If you COULD get $1500 for renting it, then it would be taxed at $750, regardless of whether you are renting it for $1000 or $1500.
Of course, assessors figuring that out is tricky.
However, your initial point I think is correct, the land tax is entirely shouldered by the owner, not the renter.
robc
Apr 27 2023 at 9:30am
nobody,
I dont think the incidence of property tax says anything about the incidence of a land tax, as most of property tax is on the improved value of the property (mostly the building), which is entirely different than a tax on unimproved land.
A Georgist Land Tax would be interesting in somewhere like the Netherlands, where the unimproved value of large chunks of the land would be the value of the sea floor, so pretty close to $0.
nobody.really
Apr 27 2023 at 1:02pm
Can a landlord “pass on” the cost of a land tax to a tenant?
Imagine the landlord could charge a market rent of $1500, but instead charges nothing. But when the landlord incurs a land tax of $750, she starts charging the tenant $750. Under robc’s interpretation, so long as the charge is no more than the rental amount presumed by the taxing authority (up to $1500), then there would be no incremental cost/tax to the landlord for charging this amount.
We might dispute whether to characterize the charge by the landlord as “passing on” the cost of the tax, or to characterize it as (below-market) rent. But I think I get the point.
Thoughtful distinction–although I can’t say whether this distinction matters for purposes of my hypothetical. But to simplify things, I’ll re-present the hypothetical using property taxes in lieu of land taxes.
vince
Apr 27 2023 at 6:28pm
The goal of the land tax is to offset economic rent. Rent on a building is theoretically a return for production rather than economic rent. Taxing return on production discourages production. Taxing land doesn’t doesn’t discourage production; land can’t be produced.
vince
Apr 26 2023 at 9:03pm
How does one even define above normal profits when it comes to innovation?
nobody.really
Apr 27 2023 at 1:09pm
[Take 2]
Consider a property tax. What if a landowner has no revenue—in part because the landowner wishes to leave her land pristine and unoccupied? I would expect the state to pursue legal claims seize any assets available. Would that constitute slavery?
If we can countenance taxing property, why not human capital?
In many states, people getting divorced must fill out worksheets that reflect how much they have been earning, and divorce decrees often direct one party to make continuous payments to another party based on an assumption that the parties’ incomes will continue. As far as I know, you have no obligation to maintain that level of income—or any level. But failure to make the prescribed support payments may result in fines, a finding of contempt of court, and jail time. Is this slavery?
And how far do we want to take this? Should the state punish parents who fail to care for their kids? Negligent parents are merely choosing how to allocate their own resources, including human resources. The idea that parents should have to expend resources on their kids is just fuzzy-headed sentimentality, right? I expect you disagree–but can you articulate a rationale?
Monte
Apr 27 2023 at 1:09pm
From your review of their book in Regulation:
As do I. But my optimism is starting to wane. If we’re to believe Paolo Gerbaudo’s claim that Big Government is Back, we can expect policy proposals designed to undercut property rights or redistribute wealth to prevail over more free market alternatives:
From the same review:
I agree that Q V is an intriguing idea. But there are drawbacks that need to be addressed, namely collusion, voter preferences polarized by wealth, and Sybil attacks.
TGGP
Apr 28 2023 at 4:20pm
I’m not generally a fan of Posner/Weyl, but I endorse taxes on inelastic features of human capital, such as this:
https://www.overcomingbias.com/p/tax-the-tallhtml
It’s similar to my reason for endorsing Georgist taxation on rent from unimproved land rather than all the destructive taxation we rely on today.
Comments are closed.