Last October I posted on Eric Posner and Glen Weyl’s proposal to generalize self-assessed property taxes. For many items, such as land and buildings, you’d pay an annual tax that is a standard percentage of your self-set sale-offer price for the item. This would avoid administrative property valuations, discourage people from sitting on stuff they don’t use, and make it much easier to assemble property into large units. Eminent domain would no longer be needed. They have a new book,
I am still not sure precisely what paper you mean, I would appreciate a more precise reference. By any chance is it the paper by Weyl and Zhang? Because in their paper, the price is transparent to the buyer, and so their analysis cannot apply directly to the new scheme.
By "prior work" do you mean Posner and Weyl's paper, or some paper cited by them, or some other paper? I couldn't find any game-theoretic analysis in Posner and Weyl. In any case, I would expect a game-theoretic analysis to allow a potential buyer to make his decision as to whether to buy an item based on the price of the item, in which case the probability of finding a buyer depends on the price. A game in which potential buyers couldn't see the price would presumably have a different game-theoretic analysis.
So then where, in your view, is the incentive for honesty (in either scheme)? I understand that there are incentives for not bidding too high or too low relative to some benchmark, but I don't understand why you expect this benchmark to be correlated with the value that someone assigns to the item / event (rather than just being a fixed property of the scheme).
As an aside, I would like to cast my vote in as one of the people who thinks your responses to comments tend to be too short (and I do not think I am the only one). I understand you don't have time to respond in detail to every comment but most of the time bare assertions are not very useful, unless the goal is to discourage participation from people who can't mind-read. I think that most academics of a similar status to you respond more to their critics than you do.
In Texas property taxes can be applied at highest and best use.
In the self-assessment scheme of property taxes, you would still have the problem of property zoning. That is if a city zones a land for only lowrise construction, then the property will be worth less.
Also, if a city has not zoned enough property as residential at enough density, you will still have housing shortages as in Hong Kong Los Angeles New York or London etc.
I wish libertarians would embrace a simple elimination of property zoning.
Difficult as it may be to believe today properties on Main was only ruled constitutional in 1926
I don't understand where the incentive for honesty is in this scheme. In the previous scheme it came from the fact that the lower you set the price of an item, the more likely it is that someone else will buy it from you. Someone who highly values the item has a larger incentive to make it unlikely that someone else will buy it than someone who values the item less highly, and consequently is incentivized to choose a larger price for the item.
However, in this new scheme, setting a low compensation for a type of event would not generally increase the probability of that event happening, unless the person who might cause the event is aware of what compensation level you have chosen. (Or maybe you are suggesting that everyone should get a decal on their car to indicate their chosen level of compensation in the case of accidents? :P ) So the damage from the event is just another sunk cost, and appears to make no difference to the calculation of what level of compensation you should choose.
Of course, I realize that my point is undermined somewhat by the concavity of utility with respect to money, and that normal insurance does work on a scheme basically like the one mentioned here (partially because of the sunk cost fallacy). But here you seem to be combining the notion of fairness in a court judgement with insurance in a way that's not usually done, and it seems to me that this means a stronger argument for why honesty is incentevized by the scheme is desirable. Also, one could argue that the concavity of utility with respect to money would only incentivize honesty with respect to financial costs.
No need to use extreme examples when there are plenty of intermediate possibilities. For example, food preparation safety in the food industry and in restaurants is regulated, rather than expecting each consumer to do their own inspections.
Limiting the domain to automobile accidents, it seems like the ability to sue for damage and buying insurance are imperfect substitutes?
To avoid the uncertainty and expense of going to court, we can buy insurance (for example, uninsured motorist insurance) to make payments in case of an accident less uncertain. Financially, this seems like a superior alternative to buying *more* rights for compensation from the other party? You'd probably rather have payouts guaranteed by some entity with proven willingness and ability to pay.
On the other hand, if you have this insurance, maybe you could save on taxes by disclaiming the right to your day in court? But the insurance company might want to preserve the right to recover costs. So, maybe the insurance company would require you to buy these rights?
Also, insurance payouts are an imperfect substitute for injury, because you get more money, but you're still injured. So, maybe you still want to buy more rights to liability for deterrence?
But, look at this from the other side: people don't choose which car to have an accident with based on that person's liability rights. Instead it happens fairly randomly. So, only the average chance of liability could act as any sort of deterrence, and there's an obvious free-rider problem here.
(Though in practice, I don't think the financial implications are the most important incentive for avoiding accidents? If you want to deter people from colliding with you, buying a stronger, heavier vehicle is probably going to do a better job of it.)
Also, it seems like there is a choice between financial certainty and risk. Getting a small, fixed income stream in exchange for taking on large, uncertain risks (for example, selling options or insurance) seems like the opposite of what most people want to do. It's only entities with large financial resources where taking on more risk is a good trade, if it pays off on average.
I'm concerned about the actual implementation vs. theory. How about coming up with ideas to simplify all/most taxes, instead of making such systems even more nuanced and complex than they are already? I fear that your proposal would make managing property taxes entail far more paperwork and computation and costly expert consultations than income taxes already do. And, no doubt, all our current income tax paperwork exists due to only the best of intentions, by the best minds (accountants and lawyers and economists and politicians and lobbyists - what could go wrong?), seeking the most just, fair, and effective tax system overall. Right? Bear in mind that these are the same people who will be writing all the laws and working out all the details and regulations in your new/improved property tax system. Perhaps this would work out well in a world populated by great and honorable minds. Or... maybe not.
Can we trust people to know their food preferences? After all, people can be pretty random; many people are not good at food assessment. Maybe we should just give everyone the same standard food package, and tax them to pay for it.
There seems to be an assumption that people will pay taxes based on a rational estimate of their preferences in hypothetical events. But in practice, such choices may be no more rational than buying lottery tickets. Judging by the risks they take (physical or financial), many people are not good at risk assessment.
Also, the consequences of available income being a major input to preferences seems not to be well-considered in these free-market schemes. It's difficult to weigh the tradeoff against all other competing needs, particularly when money is scarce.
But I wonder if there's some restricted domain where this scheme would work particularly well and the downsides not so evident? Perhaps for businesses?
The catalog only needs to be huge if people want to express a lot of detail about how their personal preferences differ from the average. If they just want to say that they dislike pain more than most people do, a couple of cases may be sufficient to express that.
You just know that there will be that one poor lady who sets the taxable valuation of her left arm unusually low, and that perverted banker who's like "Alright, I'm buying that arm! The transaction went through; my surgeons will be at your place between noon and 5pm. Another trophy for the arm chamber, my favorite place to drink red wine!"
you invented insurance
I am still not sure precisely what paper you mean, I would appreciate a more precise reference. By any chance is it the paper by Weyl and Zhang? Because in their paper, the price is transparent to the buyer, and so their analysis cannot apply directly to the new scheme.
(Edited for tone)
They cite a technical paper with math details. As I said, that math doesn't require the chance of purchase to depend on the price.
By "prior work" do you mean Posner and Weyl's paper, or some paper cited by them, or some other paper? I couldn't find any game-theoretic analysis in Posner and Weyl. In any case, I would expect a game-theoretic analysis to allow a potential buyer to make his decision as to whether to buy an item based on the price of the item, in which case the probability of finding a buyer depends on the price. A game in which potential buyers couldn't see the price would presumably have a different game-theoretic analysis.
The prior work gave explicit game theoretic analyses of equilibria. Those same analyses apply quite directly to this new application.
So then where, in your view, is the incentive for honesty (in either scheme)? I understand that there are incentives for not bidding too high or too low relative to some benchmark, but I don't understand why you expect this benchmark to be correlated with the value that someone assigns to the item / event (rather than just being a fixed property of the scheme).
As an aside, I would like to cast my vote in as one of the people who thinks your responses to comments tend to be too short (and I do not think I am the only one). I understand you don't have time to respond in detail to every comment but most of the time bare assertions are not very useful, unless the goal is to discourage participation from people who can't mind-read. I think that most academics of a similar status to you respond more to their critics than you do.
No the prior scheme didn't depend on the likelihood of purchase depending on price.
In Texas property taxes can be applied at highest and best use.
In the self-assessment scheme of property taxes, you would still have the problem of property zoning. That is if a city zones a land for only lowrise construction, then the property will be worth less.
Also, if a city has not zoned enough property as residential at enough density, you will still have housing shortages as in Hong Kong Los Angeles New York or London etc.
I wish libertarians would embrace a simple elimination of property zoning.
Difficult as it may be to believe today properties on Main was only ruled constitutional in 1926
I don't understand where the incentive for honesty is in this scheme. In the previous scheme it came from the fact that the lower you set the price of an item, the more likely it is that someone else will buy it from you. Someone who highly values the item has a larger incentive to make it unlikely that someone else will buy it than someone who values the item less highly, and consequently is incentivized to choose a larger price for the item.
However, in this new scheme, setting a low compensation for a type of event would not generally increase the probability of that event happening, unless the person who might cause the event is aware of what compensation level you have chosen. (Or maybe you are suggesting that everyone should get a decal on their car to indicate their chosen level of compensation in the case of accidents? :P ) So the damage from the event is just another sunk cost, and appears to make no difference to the calculation of what level of compensation you should choose.
Of course, I realize that my point is undermined somewhat by the concavity of utility with respect to money, and that normal insurance does work on a scheme basically like the one mentioned here (partially because of the sunk cost fallacy). But here you seem to be combining the notion of fairness in a court judgement with insurance in a way that's not usually done, and it seems to me that this means a stronger argument for why honesty is incentevized by the scheme is desirable. Also, one could argue that the concavity of utility with respect to money would only incentivize honesty with respect to financial costs.
No need to use extreme examples when there are plenty of intermediate possibilities. For example, food preparation safety in the food industry and in restaurants is regulated, rather than expecting each consumer to do their own inspections.
Limiting the domain to automobile accidents, it seems like the ability to sue for damage and buying insurance are imperfect substitutes?
To avoid the uncertainty and expense of going to court, we can buy insurance (for example, uninsured motorist insurance) to make payments in case of an accident less uncertain. Financially, this seems like a superior alternative to buying *more* rights for compensation from the other party? You'd probably rather have payouts guaranteed by some entity with proven willingness and ability to pay.
On the other hand, if you have this insurance, maybe you could save on taxes by disclaiming the right to your day in court? But the insurance company might want to preserve the right to recover costs. So, maybe the insurance company would require you to buy these rights?
Also, insurance payouts are an imperfect substitute for injury, because you get more money, but you're still injured. So, maybe you still want to buy more rights to liability for deterrence?
But, look at this from the other side: people don't choose which car to have an accident with based on that person's liability rights. Instead it happens fairly randomly. So, only the average chance of liability could act as any sort of deterrence, and there's an obvious free-rider problem here.
(Though in practice, I don't think the financial implications are the most important incentive for avoiding accidents? If you want to deter people from colliding with you, buying a stronger, heavier vehicle is probably going to do a better job of it.)
Also, it seems like there is a choice between financial certainty and risk. Getting a small, fixed income stream in exchange for taking on large, uncertain risks (for example, selling options or insurance) seems like the opposite of what most people want to do. It's only entities with large financial resources where taking on more risk is a good trade, if it pays off on average.
I'm concerned about the actual implementation vs. theory. How about coming up with ideas to simplify all/most taxes, instead of making such systems even more nuanced and complex than they are already? I fear that your proposal would make managing property taxes entail far more paperwork and computation and costly expert consultations than income taxes already do. And, no doubt, all our current income tax paperwork exists due to only the best of intentions, by the best minds (accountants and lawyers and economists and politicians and lobbyists - what could go wrong?), seeking the most just, fair, and effective tax system overall. Right? Bear in mind that these are the same people who will be writing all the laws and working out all the details and regulations in your new/improved property tax system. Perhaps this would work out well in a world populated by great and honorable minds. Or... maybe not.
Can we trust people to know their food preferences? After all, people can be pretty random; many people are not good at food assessment. Maybe we should just give everyone the same standard food package, and tax them to pay for it.
There seems to be an assumption that people will pay taxes based on a rational estimate of their preferences in hypothetical events. But in practice, such choices may be no more rational than buying lottery tickets. Judging by the risks they take (physical or financial), many people are not good at risk assessment.
Also, the consequences of available income being a major input to preferences seems not to be well-considered in these free-market schemes. It's difficult to weigh the tradeoff against all other competing needs, particularly when money is scarce.
But I wonder if there's some restricted domain where this scheme would work particularly well and the downsides not so evident? Perhaps for businesses?
I do not know that.
The catalog only needs to be huge if people want to express a lot of detail about how their personal preferences differ from the average. If they just want to say that they dislike pain more than most people do, a couple of cases may be sufficient to express that.
You just know that there will be that one poor lady who sets the taxable valuation of her left arm unusually low, and that perverted banker who's like "Alright, I'm buying that arm! The transaction went through; my surgeons will be at your place between noon and 5pm. Another trophy for the arm chamber, my favorite place to drink red wine!"