Important econ-101 point which I expect people reading the OP to be confused about: perfect competition destroys all economic profit, which is not the same as what “profit” intuitively suggests. It means that everybody has the same gains they’d have by taking some other option, not that nobody has any surplus at all. So, for instance:
Perfect competition destroys all producer surplus
That’s completely wrong if we use the usual economic notion of “surplus”—roughly speaking, at least for producers, surplus is the accounting profit (revenue minus cost). Perfect competition does not drive surplus to zero, even in the most idealized theoretical models.
What perfect competition drives to zero is economic profits. That means that, in perfect competition, producers won’t make any more profit than they would make by working in some other industry—they won’t make any excess profit. That does not mean producers won’t make any profit, in the usual accounting sense of revenue minus cost.
Which brings us to...
The worst case for consumers is if they are also the producers – a world fully given over to perfect optimization of some defined outcome. No matter how good prices of goods and quantities available might get, if you have zero surplus of any kind to spend on them, it doesn’t do you any good.
You get a Disneyland with no children.
This is completely wrong, for exactly the reason just discussed. Economic profit is not the same as profit in the usual sense. In a world of perfect competition, everybody makes exactly the same amount of profit working on any of their best options—zero economic profit, zero excess profit. They do not make zero profit, in the usual revenue minus cost sense of the word.
It would save a lot of trouble if the economists made up a different word, rather than using “economic profit”, but sadly that boat has long since sailed.
I am not making the mistake here—as Jacob bets, I am well aware that what is happening is that producers make zero economic profits, and am referring to other things. Plus zero economic profits, in current reality, is pretty close to zero profits period (after labor costs, so some profit in some sense at least), and the more the system is perfectly competitive in other potential places to invest the more economic profits equal profits and I could write a whole additional post detailing this stuff and am going to try doing that now and see if it’s worth doing.
But I can totally get how the way it is worded could lead readers, especially readers who don’t know econ terms, to get the wrong impression, so I’m going to try and reword some stuff to be more careful. I do stand by the claims as I intended them.
EDIT/update: Page is now ready for re-importing, with some expanded text to hopefully clear up and prevent these confusions and make it clear what I mean here. I also think later parts will, if successful, make a lot of it more clear. I do not think an extra post is worthwhile, but I could change that opinion later.
The Iron Law of Wages takes over, driving labor’s wage down to bare subsistence and capturing any efficiency gains from superior production. Capital demands and gets the risk-free rate, which quickly settles at or below zero.
This is still wrong—both the wage claim and the capital claim.
Iron Law of Wages doesn’t actually work unless people have more kids whenever they have spare income. This hasn’t been the case for two centuries; more income generally leads to fewer kids across most of the world today. (Though we could steelman the claim here and say that cultures which have more kids will eventually dominate the human population—then wages would drop to subsistence.)
On the capital side, I don’t know of any model at all where the risk-free rate naturally settles at zero, other than in a scenario of massive abundance. If capital is scarce at all, it’s going to settle to an above-zero rate of return in the long run.
The common theme in both cases: both wages-above-subsistence and capital returns only go to zero if the corresponding good (labor and capital, respectively) are massively abundant. Perfect competition is not sufficient for that to happen, except maybe if we mean “perfect competition between cultures” producing lots of babies.
Also...
Plus zero economic profits, in current reality, is pretty close to zero profits period (after labor costs, so some profit in some sense at least)
If we’re talking about “Disneyworld without any children”, the accounting profit made by labor is the only part that matters. Corporations making zero accounting profit (after capital cost) is fine, that’s not a problem at all, and of course that’s going to happen in any perfect competition scenario because starting a corporation is trivial in a perfectly competitive market. Corporations are massively abundant in this scenario, so of course the “price” of a corporation is zero. It’s the labor cost which is actually interesting, and it’s the laborers (including management etc) who actually get a profit, because there isn’t an unlimited supply of laborers.
Bottom line: no amount of perfect competition will make economic profits equal to actual profits. Perfect competition just isn’t a sufficient condition for that; it requires massive abundance of workers.
Like johnswentworth, I also don’t understand the leap to “Iron Law of Wages takes over”. You seem to be at least making some unstated assumptions in that part. Besides, even if subsistence wages do obtain for labor, land should continue to be very valuable and be earning a lot of rent/surplus, so can’t the landlords at least be considered “children in Disneyland”?
(I said I wasn’t going to comment further here, but for Wei_Dai, who was likely absent for a while, I’m going to make a one-time exception, but not respond further here. If Wei wants to talk further I can be contacted elsewhere.)
That was a late edit to attempt to respond to comments, and I’m already regretting it. I made a last attempt to make this more clear. If it’s still not clear, then once I’ve got to where I am eventually going, then we can revisit and see if there’s a better way to go about this.
It’s hard to know exactly which assumption is the important missing one, but my guess is it’s that we’re talking about thinking through all the implications of a fully perfectly competitive world, rather than an individual perfectly competitive market. Or, alternatively, the implicit assumption that a perfectly competitive world is static, so we are solving for the equilibrium, although I try to avoid pulling that trump card out at steps where one doesn’t have to.
The labor market is fully competitive by assumption. Or, given undifferentiated labor, the only way to not have the Iron Law be true right away (other than frictions we are assuming away) is to have a labor shortage because demand for goods exceeds labor available to supply the goods. Now the question comes of where that excess demand is coming from.
On land: I understand in its historical context why land is distinct from capital in this kind of context, and it certainly can have different dynamics in some ways, but I’ve always wondered why land isn’t just capital like any other capital, once it is properly valued. If we agree to assume the risk-free rate truly is zero, then the value of land is equal to the sum of its future income streams, which if things truly are permanently static means land is infinitely expensive—extremes do insane things. More realistically, land is just super expensive (see: SF, NYC, etc, only more so) so landlords have very high wealth, which they can spend for at least some time, but this doesn’t ‘get us out’ any more than other sources of wealthy people get us temporarily out for that subset, which must be small.
(You could also say that land being valuable is actually a violation of perfect competition since it implies differentiation in costs, again extreme situations are weird.)
I’ve always wondered why land isn’t just capital like any other capital, once it is properly valued. If we agree to assume the risk-free rate truly is zero...
This has never, ever been the case in the history of humanity. The risk-free rate of land is the one thing we actually have very good records for and ways of calculating over human history. It’s not zero, nor has it ever been close to zero. Somewhere around 3-5% on average is a reasonably good estimate without getting into specifics of region and era.
But if you’re basing your argument on land having a risk free return on zero, we’ll that’s something that can be quite easily looked up and disproven.
(This will be my last word in this thread, as I do think replying to this can be helpful in creating clarity, then I’m done here.)
The argument here is not based on ’there was a time and place, or is a time and place, where this scenario took place. It is an argument that given the assumptions, this is what would happen. It definitely hasn’t happened in the past.
Agreed that the rate of return on land has never been zero until at least recently because returns to capital have not been zero until recently if ever.
But if returns to non-land capital are zero, and returns to land capital are positive, and land is alienable, then people bid up land until the returns equalize. If it’s not alienable, as it sometimes isn’t, asking what it is worth is weird but I’d assume you want to value it as if it was alienable anyway.
This brings me back to not understanding the land vs. non-land division absent laws about land ownership. Land is a capital good like any other, its return must be the same as all the others unless something is out of equilibrium.
All right, that’s it. I sincerely wish I’d been able to convey the points in ways that didn’t lead to these confusions, or found ways to skip over the confusing things and still get where I want to go, but what’s done is done.
Given this (which I’d bet Zvi is well aware of), I’m quite confused about what the Disneyland sentence is supposed to mean; and I’d be curious for Zvi to clarify.
Important econ-101 point which I expect people reading the OP to be confused about: perfect competition destroys all economic profit, which is not the same as what “profit” intuitively suggests. It means that everybody has the same gains they’d have by taking some other option, not that nobody has any surplus at all. So, for instance:
That’s completely wrong if we use the usual economic notion of “surplus”—roughly speaking, at least for producers, surplus is the accounting profit (revenue minus cost). Perfect competition does not drive surplus to zero, even in the most idealized theoretical models.
What perfect competition drives to zero is economic profits. That means that, in perfect competition, producers won’t make any more profit than they would make by working in some other industry—they won’t make any excess profit. That does not mean producers won’t make any profit, in the usual accounting sense of revenue minus cost.
Which brings us to...
This is completely wrong, for exactly the reason just discussed. Economic profit is not the same as profit in the usual sense. In a world of perfect competition, everybody makes exactly the same amount of profit working on any of their best options—zero economic profit, zero excess profit. They do not make zero profit, in the usual revenue minus cost sense of the word.
It would save a lot of trouble if the economists made up a different word, rather than using “economic profit”, but sadly that boat has long since sailed.
I am not making the mistake here—as Jacob bets, I am well aware that what is happening is that producers make zero economic profits, and am referring to other things. Plus zero economic profits, in current reality, is pretty close to zero profits period (after labor costs, so some profit in some sense at least), and the more the system is perfectly competitive in other potential places to invest the more economic profits equal profits and I could write a whole additional post detailing this stuff and am going to try doing that now and see if it’s worth doing.
But I can totally get how the way it is worded could lead readers, especially readers who don’t know econ terms, to get the wrong impression, so I’m going to try and reword some stuff to be more careful. I do stand by the claims as I intended them.
EDIT/update: Page is now ready for re-importing, with some expanded text to hopefully clear up and prevent these confusions and make it clear what I mean here. I also think later parts will, if successful, make a lot of it more clear. I do not think an extra post is worthwhile, but I could change that opinion later.
So I see this update:
This is still wrong—both the wage claim and the capital claim.
Iron Law of Wages doesn’t actually work unless people have more kids whenever they have spare income. This hasn’t been the case for two centuries; more income generally leads to fewer kids across most of the world today. (Though we could steelman the claim here and say that cultures which have more kids will eventually dominate the human population—then wages would drop to subsistence.)
On the capital side, I don’t know of any model at all where the risk-free rate naturally settles at zero, other than in a scenario of massive abundance. If capital is scarce at all, it’s going to settle to an above-zero rate of return in the long run.
The common theme in both cases: both wages-above-subsistence and capital returns only go to zero if the corresponding good (labor and capital, respectively) are massively abundant. Perfect competition is not sufficient for that to happen, except maybe if we mean “perfect competition between cultures” producing lots of babies.
Also...
If we’re talking about “Disneyworld without any children”, the accounting profit made by labor is the only part that matters. Corporations making zero accounting profit (after capital cost) is fine, that’s not a problem at all, and of course that’s going to happen in any perfect competition scenario because starting a corporation is trivial in a perfectly competitive market. Corporations are massively abundant in this scenario, so of course the “price” of a corporation is zero. It’s the labor cost which is actually interesting, and it’s the laborers (including management etc) who actually get a profit, because there isn’t an unlimited supply of laborers.
Bottom line: no amount of perfect competition will make economic profits equal to actual profits. Perfect competition just isn’t a sufficient condition for that; it requires massive abundance of workers.
Like johnswentworth, I also don’t understand the leap to “Iron Law of Wages takes over”. You seem to be at least making some unstated assumptions in that part. Besides, even if subsistence wages do obtain for labor, land should continue to be very valuable and be earning a lot of rent/surplus, so can’t the landlords at least be considered “children in Disneyland”?
(I said I wasn’t going to comment further here, but for Wei_Dai, who was likely absent for a while, I’m going to make a one-time exception, but not respond further here. If Wei wants to talk further I can be contacted elsewhere.)
That was a late edit to attempt to respond to comments, and I’m already regretting it. I made a last attempt to make this more clear. If it’s still not clear, then once I’ve got to where I am eventually going, then we can revisit and see if there’s a better way to go about this.
It’s hard to know exactly which assumption is the important missing one, but my guess is it’s that we’re talking about thinking through all the implications of a fully perfectly competitive world, rather than an individual perfectly competitive market. Or, alternatively, the implicit assumption that a perfectly competitive world is static, so we are solving for the equilibrium, although I try to avoid pulling that trump card out at steps where one doesn’t have to.
The labor market is fully competitive by assumption. Or, given undifferentiated labor, the only way to not have the Iron Law be true right away (other than frictions we are assuming away) is to have a labor shortage because demand for goods exceeds labor available to supply the goods. Now the question comes of where that excess demand is coming from.
On land: I understand in its historical context why land is distinct from capital in this kind of context, and it certainly can have different dynamics in some ways, but I’ve always wondered why land isn’t just capital like any other capital, once it is properly valued. If we agree to assume the risk-free rate truly is zero, then the value of land is equal to the sum of its future income streams, which if things truly are permanently static means land is infinitely expensive—extremes do insane things. More realistically, land is just super expensive (see: SF, NYC, etc, only more so) so landlords have very high wealth, which they can spend for at least some time, but this doesn’t ‘get us out’ any more than other sources of wealthy people get us temporarily out for that subset, which must be small.
(You could also say that land being valuable is actually a violation of perfect competition since it implies differentiation in costs, again extreme situations are weird.)
This has never, ever been the case in the history of humanity. The risk-free rate of land is the one thing we actually have very good records for and ways of calculating over human history. It’s not zero, nor has it ever been close to zero. Somewhere around 3-5% on average is a reasonably good estimate without getting into specifics of region and era.
But if you’re basing your argument on land having a risk free return on zero, we’ll that’s something that can be quite easily looked up and disproven.
(This will be my last word in this thread, as I do think replying to this can be helpful in creating clarity, then I’m done here.)
The argument here is not based on ’there was a time and place, or is a time and place, where this scenario took place. It is an argument that given the assumptions, this is what would happen. It definitely hasn’t happened in the past.
Agreed that the rate of return on land has never been zero until at least recently because returns to capital have not been zero until recently if ever.
But if returns to non-land capital are zero, and returns to land capital are positive, and land is alienable, then people bid up land until the returns equalize. If it’s not alienable, as it sometimes isn’t, asking what it is worth is weird but I’d assume you want to value it as if it was alienable anyway.
This brings me back to not understanding the land vs. non-land division absent laws about land ownership. Land is a capital good like any other, its return must be the same as all the others unless something is out of equilibrium.
All right, that’s it. I sincerely wish I’d been able to convey the points in ways that didn’t lead to these confusions, or found ways to skip over the confusing things and still get where I want to go, but what’s done is done.
Given this (which I’d bet Zvi is well aware of), I’m quite confused about what the Disneyland sentence is supposed to mean; and I’d be curious for Zvi to clarify.