Good blog. The theory of negative externalities is really silly if not supplemented by an ethical theory which makes clear which externalities may or may not be taxed. It is the same reasoning with public goods too. Pretty girls in public streets are positive externalities, yet I doubt many would advocate to subsidies them going out to ensure the optimum supply. Or tax the obese and ugly to keep them inside.
I agree, but that still understates it. As poor as the measurement and justification of a reaction to an externality may be, there is no mechanism to try to measure the private costs of a tax or subsidy. At best it is incomplete. We can at least think of a model of market behavior that urges people to prioritize what is most important to them. Is there anything similar when it comes to externality?
To be fair the market already rewards attractiveness. For women, and only for women, a healthy body weight is correlated with a higher salary, because male (and possibly also female?) bosses will hire or promote hot workers, all other things being equal.
Well, sure, but that only captures the benefit to people who interact economically with them. The positive externality is *on top of that*, to the other people who get to see them in public places for free.
I think the explanation for why governments don't apply the appropriate Pigovian subsidies and penalties comes down to political practicality. A negative income tax or a subsidy helping older Americans or ppl affected by covid to keep living in their apartments would require substantial taxation that would be very salient to most voters. OTOH pass a minimum wage law or a rent control ordinance and the costs are hidden and non-salient.
Indeed, I think this is a huge part of the reason the left is so opposed to free market mechanisms. Not because they have a theoretical objection to the incentives based approach but because they know it's politically implausible and thus advocating for it isn't an honest signal of concern (people can say they support it with little fear it will ever happen).
I'm a bit confused of the relevance. We borrowed for COVID era and it was temporary and extraordinary so I'm not clear on how it conflicts with my theory.
Good stuff. The first thing that popped into my head was that catalytic converters could have been handled in this or some similar way. Tax a polluting car, or install a catalytic converter in lieu of paying the tax. You could have done seatbelts the same way.
There are genuine cases of oberbearing government regulation, but seatbelts ain't it, chief. Nobody is complaining about it being impossible to buy a seatbelt-less car, mandatory seatbelts are at worst a neutral policy if you take the (not unreasonable) view that there wouldn't be a market for cars without seatbelts even if they were legal.
One argument against Pigovian solutions is that they’re often simply not practical. How do you assess a tax on the external cost of dumping waste into a river? Or determine the external value of planting a garden in front of your house? In practice, it’s likely the case that for many externality-imposing activities, either almost no one would partake at the adjusted cost, as it would be way too high to be worth it, or everyone would, because the tax would be trivial. The cost of going through he rigmarole of assessing the value/cost of the externality and collecting the tax may not be worth the slight efficiency gain over just imposing a straightforward ban.
When i try to explain this principle of subsidizing or taxing externalites, but then leaving businesses alone, a common response is “so we should just allow businesses to do bad things!? This tax/subsidy punishes normal people, why not just ban the bad thing??”
People usually think in very black and white terms: trade offs or odds is not something normal people want to think about.
Why don't these people realise that banning the bad thing also punishes ordinary people, worse than merely taxing it? Isn't it obvious that, from the point of view of a drinker, alcohol duty is bad but prohibition would be even worse?
There's been an argument recently that the government isn't capable of such fine-grained policy, and that only blunt policies like prohibition are viable. I wish I could fine the Twitter thread now, but it was remarking on the recent experience with marijuana decriminalization. Legal, regulated & taxed marijuana businesses are still being vastly outsold by completely illegal ones, and there's basically no enforcement to get the criminals to start obeying the law.
While I agree that it does come down to practicality I don't think it's legal practicality. For the most part, cutting checks is pretty easy (taxing negative externalities is a bit harder) but the problem is that those taxes would be very visible and salient to voters while we don't see the costs that we pay because of, say, a minimum wage.
Re: MJ I don't think the dispensaries are being outsold. Rather, it's that the bulk supply of MJ even for them comes from the illegal market.
But that's because of federal illegality and the fact that the laws were delibrately written with an unworkable hippie vision of small growers completely ignoring the returns to scale and the huge fixed costs of running a federally illegal buisness w/o access to banking.
One could at least make an argument that a tax could be applied in a way that penalizes only those who benefit from a negative externality. But I don’t think there is much possibility of basing a subsidy on taxes that fall only on gainers. And the first tax fails to compensate those who bear the cost of the negative externality.
It all depends on bureaucrats being motivated and able to measure things that are hard to measure and potentially embarrassing. How good does the estimate need to be before it justifies coercion?
I don't think it does. It merely requires that bureaucrats not be much worse at assessing and applying taxes and subsidies than they are at applying prohibitions. If they are bad at figuring out when and who to tax they aren't going to be better at applying the blunter tool of prohibitions.
And it's hard to see how they could be given that, in some sense, a prohibition is really just an infinite tax (or, alternatively, anything you can forbid you can say you can do this if you pay X amount).
I think the issue is much more political feasibility. There isn't much of a question that from an economic and fairness POV it would be superior to replace the minimum wage with a negative income tax in a way that ensures everyone had the chance to get a decent wage. It wouldn't be hard (government is relatively good at the whole tax and checks thing) and it would have less non-compliance because no one has an incentive to make banned off book deals.
It doesn't happen because you'd see that money coming out of your paycheck in taxes while you don't see the higher prices you pay for goods because of the minimum wage.
I think you missed my point. Given a formula for who pays how much, I have no doubt it could be collected. But how do they come up with a formula appropriate to the objective of the tax or subsidy? I am not filled with confidence.
These ideas have been around for a long time. I’m not sure whether they don’t get much practical use because they are hard to implement correctly or because no one really has an incentive to use them.
Suppose they just try to estimate it and often don't get it exactly right? Why is it going to be a worse fit policy than a prohibition? It's not that it's an easy problem to solve, it's just that you don't avoid it by selecting the prohibition method (why is that prohibition the appropriate degree of incentive).
Or, to put the point differently, a prohibition is just setting the penalty for doing something exceptionally high. Sometimes so high that it requires prison time but often when regulating buisness it's a punitive fine. Why is setting a really high penalty (fine) for a certain behavior any more likely to be the optimal (or just good) solution than picking a tax rate.
We just have a general problem of designing incentives that maximize social welfare. Sometimes it's going to require really large penalties (murder) but other times we have good theoretical reason to think that a better outcome would be achieved via less extreme incentives (taxes) and also the background belief that we should generally try not to forbid things we don't have to. In those cases why not try the Pigovian tax model first and if we see evidence it's socially non-optimal it's alot easier to tweak the rates (and easier to get useful data about effects of tweaks) than it is to tweak hard prohibitions.
“ if we see evidence it's socially non-optimal it's alot easier to tweak the rates”
My point is that what is socially optimal is not clear or unanimous, and they don’t have much incentive to find it if it exists.
It is subjective and collective. One person's evaluation is not the same as others'. What proxy should they use? It is not analogous to an exchange, where it is reasonable to expect the persons involved to have a good idea whether it makes them better off or not, and the trade-offs involved.
I think your answer overlooks the difference between what I presume Pigou had in mind and the basis for criminal law. I conceptualize Pigou as imagining a way for winners to compensate losers when they evaluate cooperative outcomes differently. It is hard to think of murderers and thieves as compensating their victims by going to jail. How much would someone have to offer me for me to consent to being killed? Prohibitions are not primarily about compensation. They are about everyone agreeing not to do to others what they don’t want happening to themselves and those they care about, and enforcing that with penalties that are proportionate to the harm. Thinking of it as a sort of price overlooks some important factors - most importantly, the involuntary and uncooperative nature of the interaction. Pigou was trying to find a way for people to cooperate, criminal law seeks a way to deal with outright defection.
I am probably projecting some of Coase's ideas into Pigou's analysis.
A huge problem is that attempts to mitigate externalities over the course of a couple centuries has created too many dials, each one adjusting something slightly different, and sometimes each other, creating an economic policy situation so complex that nobody can actually get a handle on what's going on or what to do. And we're stuck that way because for every dial, there are people with a vested personal interest in the existence of the dial and what it's set to.
In "Capitalism," George Reisman has a thoughtful discussion of externalities, both positive and negative, arguing that there is no basis for a government remedy of alleged negative externalities. See "Capitalism, A Treatise on Economics," page 1034, 'Environmentalism and the Externalities Doctrine.' I believe this brief (four page) passage might be beneficial to your class discussion.
In _Markets Dont Fail_, Brian Simpson denies externalities in principle, saying that economics considers only effects on private property. And that externalities is a rationalization of altruism.
That's just definitional fudging. You can call it what you want but most people do care about things like overall social utility so it really isn't relevant if it's altruism or not. Still the case that emitting lots of pollution makes lives worse.
Social approval does not create reality. Definitions are a need of the minds functioning. The benefits of fossil fuels from irrigation, heating, transportation, communication, cooling, cooking, etc,etc. vastly outweigh the problems. Of course, anti-[fill in the blank] Leftists hate man so benefits are irrelevant to them.
> From the perspective of the externalities doctrine, it is a flaw of capitalism whenever an individual’s action provides any kind of benefits to others for which he is not compensated, or imposes any kind of costs on others for which he does not compensate them.
-Reisman
A beautiful woman walked near me recently. Mumbling about externaIities, I tried to give her $10. She slapped my face. Should I put that money in an escrow account for a future lawsuit?
> instead of following the questions to their logical end they spew out whether ever is the status quo in their political group?
Thats the complaint of Hazlitt and Bastiat. The use of statistics for short-range economics and policies is impressive to people who accept arbitrary and conventional ideas. 51.94809 units of X seems mighty scientific to people who evade studying X.
Good blog. The theory of negative externalities is really silly if not supplemented by an ethical theory which makes clear which externalities may or may not be taxed. It is the same reasoning with public goods too. Pretty girls in public streets are positive externalities, yet I doubt many would advocate to subsidies them going out to ensure the optimum supply. Or tax the obese and ugly to keep them inside.
I agree, but that still understates it. As poor as the measurement and justification of a reaction to an externality may be, there is no mechanism to try to measure the private costs of a tax or subsidy. At best it is incomplete. We can at least think of a model of market behavior that urges people to prioritize what is most important to them. Is there anything similar when it comes to externality?
To be fair the market already rewards attractiveness. For women, and only for women, a healthy body weight is correlated with a higher salary, because male (and possibly also female?) bosses will hire or promote hot workers, all other things being equal.
Well, sure, but that only captures the benefit to people who interact economically with them. The positive externality is *on top of that*, to the other people who get to see them in public places for free.
I think the explanation for why governments don't apply the appropriate Pigovian subsidies and penalties comes down to political practicality. A negative income tax or a subsidy helping older Americans or ppl affected by covid to keep living in their apartments would require substantial taxation that would be very salient to most voters. OTOH pass a minimum wage law or a rent control ordinance and the costs are hidden and non-salient.
Indeed, I think this is a huge part of the reason the left is so opposed to free market mechanisms. Not because they have a theoretical objection to the incentives based approach but because they know it's politically implausible and thus advocating for it isn't an honest signal of concern (people can say they support it with little fear it will ever happen).
Most western countries had furlough schemes though. We did have enormous subsidies in the COVID era (rightly).
I'm a bit confused of the relevance. We borrowed for COVID era and it was temporary and extraordinary so I'm not clear on how it conflicts with my theory.
Good stuff. The first thing that popped into my head was that catalytic converters could have been handled in this or some similar way. Tax a polluting car, or install a catalytic converter in lieu of paying the tax. You could have done seatbelts the same way.
There are genuine cases of oberbearing government regulation, but seatbelts ain't it, chief. Nobody is complaining about it being impossible to buy a seatbelt-less car, mandatory seatbelts are at worst a neutral policy if you take the (not unreasonable) view that there wouldn't be a market for cars without seatbelts even if they were legal.
Well, you pay a finite for not having either.
One argument against Pigovian solutions is that they’re often simply not practical. How do you assess a tax on the external cost of dumping waste into a river? Or determine the external value of planting a garden in front of your house? In practice, it’s likely the case that for many externality-imposing activities, either almost no one would partake at the adjusted cost, as it would be way too high to be worth it, or everyone would, because the tax would be trivial. The cost of going through he rigmarole of assessing the value/cost of the externality and collecting the tax may not be worth the slight efficiency gain over just imposing a straightforward ban.
Great post
When i try to explain this principle of subsidizing or taxing externalites, but then leaving businesses alone, a common response is “so we should just allow businesses to do bad things!? This tax/subsidy punishes normal people, why not just ban the bad thing??”
People usually think in very black and white terms: trade offs or odds is not something normal people want to think about.
Why don't these people realise that banning the bad thing also punishes ordinary people, worse than merely taxing it? Isn't it obvious that, from the point of view of a drinker, alcohol duty is bad but prohibition would be even worse?
There's been an argument recently that the government isn't capable of such fine-grained policy, and that only blunt policies like prohibition are viable. I wish I could fine the Twitter thread now, but it was remarking on the recent experience with marijuana decriminalization. Legal, regulated & taxed marijuana businesses are still being vastly outsold by completely illegal ones, and there's basically no enforcement to get the criminals to start obeying the law.
While I agree that it does come down to practicality I don't think it's legal practicality. For the most part, cutting checks is pretty easy (taxing negative externalities is a bit harder) but the problem is that those taxes would be very visible and salient to voters while we don't see the costs that we pay because of, say, a minimum wage.
Re: MJ I don't think the dispensaries are being outsold. Rather, it's that the bulk supply of MJ even for them comes from the illegal market.
But that's because of federal illegality and the fact that the laws were delibrately written with an unworkable hippie vision of small growers completely ignoring the returns to scale and the huge fixed costs of running a federally illegal buisness w/o access to banking.
One could at least make an argument that a tax could be applied in a way that penalizes only those who benefit from a negative externality. But I don’t think there is much possibility of basing a subsidy on taxes that fall only on gainers. And the first tax fails to compensate those who bear the cost of the negative externality.
It all depends on bureaucrats being motivated and able to measure things that are hard to measure and potentially embarrassing. How good does the estimate need to be before it justifies coercion?
I don't think it does. It merely requires that bureaucrats not be much worse at assessing and applying taxes and subsidies than they are at applying prohibitions. If they are bad at figuring out when and who to tax they aren't going to be better at applying the blunter tool of prohibitions.
And it's hard to see how they could be given that, in some sense, a prohibition is really just an infinite tax (or, alternatively, anything you can forbid you can say you can do this if you pay X amount).
I think the issue is much more political feasibility. There isn't much of a question that from an economic and fairness POV it would be superior to replace the minimum wage with a negative income tax in a way that ensures everyone had the chance to get a decent wage. It wouldn't be hard (government is relatively good at the whole tax and checks thing) and it would have less non-compliance because no one has an incentive to make banned off book deals.
It doesn't happen because you'd see that money coming out of your paycheck in taxes while you don't see the higher prices you pay for goods because of the minimum wage.
I think you missed my point. Given a formula for who pays how much, I have no doubt it could be collected. But how do they come up with a formula appropriate to the objective of the tax or subsidy? I am not filled with confidence.
These ideas have been around for a long time. I’m not sure whether they don’t get much practical use because they are hard to implement correctly or because no one really has an incentive to use them.
Suppose they just try to estimate it and often don't get it exactly right? Why is it going to be a worse fit policy than a prohibition? It's not that it's an easy problem to solve, it's just that you don't avoid it by selecting the prohibition method (why is that prohibition the appropriate degree of incentive).
Or, to put the point differently, a prohibition is just setting the penalty for doing something exceptionally high. Sometimes so high that it requires prison time but often when regulating buisness it's a punitive fine. Why is setting a really high penalty (fine) for a certain behavior any more likely to be the optimal (or just good) solution than picking a tax rate.
We just have a general problem of designing incentives that maximize social welfare. Sometimes it's going to require really large penalties (murder) but other times we have good theoretical reason to think that a better outcome would be achieved via less extreme incentives (taxes) and also the background belief that we should generally try not to forbid things we don't have to. In those cases why not try the Pigovian tax model first and if we see evidence it's socially non-optimal it's alot easier to tweak the rates (and easier to get useful data about effects of tweaks) than it is to tweak hard prohibitions.
“ if we see evidence it's socially non-optimal it's alot easier to tweak the rates”
My point is that what is socially optimal is not clear or unanimous, and they don’t have much incentive to find it if it exists.
It is subjective and collective. One person's evaluation is not the same as others'. What proxy should they use? It is not analogous to an exchange, where it is reasonable to expect the persons involved to have a good idea whether it makes them better off or not, and the trade-offs involved.
I think your answer overlooks the difference between what I presume Pigou had in mind and the basis for criminal law. I conceptualize Pigou as imagining a way for winners to compensate losers when they evaluate cooperative outcomes differently. It is hard to think of murderers and thieves as compensating their victims by going to jail. How much would someone have to offer me for me to consent to being killed? Prohibitions are not primarily about compensation. They are about everyone agreeing not to do to others what they don’t want happening to themselves and those they care about, and enforcing that with penalties that are proportionate to the harm. Thinking of it as a sort of price overlooks some important factors - most importantly, the involuntary and uncooperative nature of the interaction. Pigou was trying to find a way for people to cooperate, criminal law seeks a way to deal with outright defection.
I am probably projecting some of Coase's ideas into Pigou's analysis.
A huge problem is that attempts to mitigate externalities over the course of a couple centuries has created too many dials, each one adjusting something slightly different, and sometimes each other, creating an economic policy situation so complex that nobody can actually get a handle on what's going on or what to do. And we're stuck that way because for every dial, there are people with a vested personal interest in the existence of the dial and what it's set to.
Or perhaps the theory of externalities is just a fig leaf covering what “public servants” really have in mind?
In "Capitalism," George Reisman has a thoughtful discussion of externalities, both positive and negative, arguing that there is no basis for a government remedy of alleged negative externalities. See "Capitalism, A Treatise on Economics," page 1034, 'Environmentalism and the Externalities Doctrine.' I believe this brief (four page) passage might be beneficial to your class discussion.
Sorry, I should have said page 96, "Environmentalism and the Externalities Doctrine," Reisman's critique of externalities.
In _Markets Dont Fail_, Brian Simpson denies externalities in principle, saying that economics considers only effects on private property. And that externalities is a rationalization of altruism.
That's just definitional fudging. You can call it what you want but most people do care about things like overall social utility so it really isn't relevant if it's altruism or not. Still the case that emitting lots of pollution makes lives worse.
Social approval does not create reality. Definitions are a need of the minds functioning. The benefits of fossil fuels from irrigation, heating, transportation, communication, cooling, cooking, etc,etc. vastly outweigh the problems. Of course, anti-[fill in the blank] Leftists hate man so benefits are irrelevant to them.
Reisman as well, page 1034, "Capitalism."
What, exactly, on p1034?
p 1034 is "p" in the index.
Sorry, I should have said page 96, "Environmentalism and the Externalities Doctrine," Reisman's critique of externalities.
> From the perspective of the externalities doctrine, it is a flaw of capitalism whenever an individual’s action provides any kind of benefits to others for which he is not compensated, or imposes any kind of costs on others for which he does not compensate them.
-Reisman
A beautiful woman walked near me recently. Mumbling about externaIities, I tried to give her $10. She slapped my face. Should I put that money in an escrow account for a future lawsuit?
> instead of following the questions to their logical end they spew out whether ever is the status quo in their political group?
Thats the complaint of Hazlitt and Bastiat. The use of statistics for short-range economics and policies is impressive to people who accept arbitrary and conventional ideas. 51.94809 units of X seems mighty scientific to people who evade studying X.