This potentially seems related to something I noticed in a post I wrote about futarchies that change their preferences: Research into how to better achieve the goals of the futarchy is something that can potentially be rewarded by the futarchy itself.
In allowing the amendment period to prevent situations where people spend 15 on research to increase the pie from 100 to 105 while transferring 16 to themselves, we simultaneously prevent people from transferring to themselves in cases where research is worth it. In the good situation where someone can spend 1 on research to increase the pie by 5, they now won't do so, since they won't see the gains, even though doing so would have been net good, even if they did bundle 1-3 transfers to themselves. Perhaps we could prevent a maximally destructive transfer in some other way, like by imposing a value on fairness in the futarchy target.
I said to pay the proposer a fraction of the market estimated welfare gain. So if that is 5, going from 100 to 105, a 25% cut would pay them 1.25, more than your assumed cost of 1.
Do you have any evidence that amendments to a bill are more likely to solve the transfer problem as opposed to creating new transfer problems? We often hear about riders on a bill that promote some special interest unrelated to the main bill, and these riders are often attached via amendment.
It's not clear to me what the difference would be between proposing an amendment in futarchy and just proposing a new bill. So your special interest group proposes something that gains +16 for them, -20 for others, and bundles it with something that gives +5 to everybody. If someone notices this, why don't they just, immediately afterwards, propose a bill that reverses the +16 -20 part of the first bill? That proposal would be +4 on net and would pass.
I don't see where you said this. You said amendments could be proposed if the proposer pays a set fee, and you said bills could be proposed if the proposer wins an auction. Which is easier to propose depends on whether the set fee is higher or lower than the auction price, which you didn't say.
But let's say the set fee is lower. Then, how is an amendment different from just a cheaper bill? Why not propose an amendment that tosses out the entire text of the bill and replaces it with a new one?
I also don't see where you said anything about a reward being paid to the proposer, other than the indirect (presumably illicit) reward if the proposer's policy proposal includes transfers to the proposer. Of course the proposer also stands to gain if they bet on their policy improving total utility and are right, but I don't see how that's any different between a bill and an amendment either.
Here is where I spoke of a reward: "The winner submits a bill, pays their auction price, and will be paid some share of the market estimated value gain if futarchy approves some variation on their bill"
Amendments are more limited in scope, cheaper to propose, and get paid lower rewards.
Limited in scope how and by whom? Whether a part of a bill is a transfer that does not increase societal welfare is a very difficult thing to objectively evaluate. A judge determining this has to determine what does and does not increase societal welfare, which is what the prediction market is supposed to do.
As your penultimate paragraph indicates, relying less on plain text for policy contents and rather on a well-structured parameter space opens up mathematical analysis and search for optimality. Basically, policy-as-code.
A perhaps less ambitious but more immediately amenable solution is to impose a system-level tax on transfers complexity: the more transfers, the higher the tax to be paid by the proposer (this is the most naive version). Similarly to eg what the ethereum virtual machine imposes through gas costs.
Why do we assume that there’s a tradeoff between redistribution (pie share transfers) and growth (changing the size of the pie)? I understand why this theoretically makes under many orthodox economic theories but I’m not convinced it’s empirically true. At least in American economic history, it seems like the highest levels of growth happened in the post-war era where there was significant government intervention in the economy and redistribution via taxes and unions, etc.
I understand that this correlation doesn’t imply causation, but it seems like the onus should be on free marketers to prove why their system is more efficient, not the other way around.
I think you're misunderstanding the transfer problem. The transfer problem is not just "any time the govt transfers wealth from person A to person B." It's specifically when the govt does this in a way that does not increase the total utility. It can be very good for total utility to redistribute wealth from the richest 1% to the people, because a dollar makes a much bigger improvement to the life of a poor person than a rich person. That's not the transfer problem.
Really it can be the transfer problem when the govt reverses this - tax cuts are the transfer problem. They benefit a special interest group (the rich) at the cost of total utility.
Most taxes cause deadweight loss. Pigovian taxes are an exception which increase utility, but those are considered "regressive" precisely because they aren't indexed to income. In order to efficiently raise lots of revenue, you need a broad base of taxation, such as via VAT, which again is regarded as more regressive than most federal taxes in the US.
The marginal value of a dollar is much higher for a poor person than for a rich person. If we consider goods assignment, it is less efficient for a billionaire to have a 5th house costing $10 million, than for a thousand poor, struggling people to have $10,000 worth of staple goods each; each individual poor person will each experience a far greater change in happiness than the billionaire will, and there are a thousand poor people to only one billionaire. This is not a small effect.
Deadweight loss is in units of (Price * Quantity). But societal welfare is in units of (Happiness * Quantity). The happiness caused by a dollar worth of personal goods is strongly dependent on how much money that person already has. So it is beneficial to accept a large amount of deadweight loss, if doing so results in production shifting towards goods that benefit poorer people instead of richer people.
"Happiness" is not a unit. You don't have actual measurements of it. And the distribution of goods is not random, rather diminishing marginal utility affects the actual decisions people make in acquiring them. We can infer its existence from such decisions. It is to the "revealed preference" of decisions we can look to rather than theorizing a priori about people's utility functions.
Also, "revealed preference" of purchasing decisions is weighted by money. For a poor person, spending $10,000 reveals a very strong preference or need for something. For Jeff Bezos, it reveals nothing more than mild interest; that amount of money is nothing to him. The amount of money a person spends on things is not an accurate reflection of how much those things are preferred, unless you somehow correct for the amount of money the person has.
Happiness can be measured (somewhat inaccurately) with polls, and of course we have a lot of common sense understanding of it.
It's hard to measure happiness but no one doubts it exists. An ideal society should be maximizing some weighted metric of happiness, health, education, scientific advance, security.
Also, deadweight loss applies only when the government is taxing a business for each unit of goods produced. If we are instead speaking of a personal income or wealth tax, the deadweight loss diagram does not apply, because an individual's decisions about how hard to work are not strongly dependent on how much they are paid. A billionaire does not work a thousand times harder than a millionaire. Beyond a certain point, money and the luxury goods bought with it are just a way of keeping score, and people care more about making more money than their peer group than they do about having any particular absolute amount. If a peer group of rich people each have their income cut in half, they would still act in much the same way - same as if you cut the number of points scored by each basket in a basketball game in half, the players still do the same things.
(Rather unthinkable to cut a group of rich people's income in half - but what are the numbers on how many times the richest people's incomes have multiplied, inflation-adjusted, since the 80s? It's a lot more than 2. And yet their behavior hasn't changed much.)
Education is not an end, but a means. And when it largely consists in signalling, it collectively achieves little.
> If we are instead speaking of a personal income or wealth tax, the deadweight loss diagram does not apply, because an individual's decisions about how hard to work are not strongly dependent on how much they are paid.
You don't know what you're talking about. The decade with the greatest increase of women into the labor force was the 1980s, because of cuts to marginal tax rates. Women were marginal workers. People make decisions about entering or leaving the labor force (including how long they stay in school or when to retire), how many hours to work if they are part time, which jobs to accept when they are looking for work, etc.
> A billionaire does not work a thousand times harder than a millionaire.
In what units are you measuring how "hard" someone works?
> If a peer group of rich people each have their income cut in half, they would still act in much the same way - same as if you cut the number of points scored by each basket in a basketball game in half, the players still do the same things.
This blog post is not about AI. And if you're suggesting I'm AI, note that I've been commenting on OB since it started in the '00s. I've also been blogging at https://entitledtoanopinion.wordpress.com/ since 2007.
Of course, I was speculating that 'Frank' was perhaps AI or AI driven and meant to post somewhere else but got mixed up somehow. Wasn't accusing you, sorry.
This potentially seems related to something I noticed in a post I wrote about futarchies that change their preferences: Research into how to better achieve the goals of the futarchy is something that can potentially be rewarded by the futarchy itself.
In allowing the amendment period to prevent situations where people spend 15 on research to increase the pie from 100 to 105 while transferring 16 to themselves, we simultaneously prevent people from transferring to themselves in cases where research is worth it. In the good situation where someone can spend 1 on research to increase the pie by 5, they now won't do so, since they won't see the gains, even though doing so would have been net good, even if they did bundle 1-3 transfers to themselves. Perhaps we could prevent a maximally destructive transfer in some other way, like by imposing a value on fairness in the futarchy target.
I said to pay the proposer a fraction of the market estimated welfare gain. So if that is 5, going from 100 to 105, a 25% cut would pay them 1.25, more than your assumed cost of 1.
Do you have any evidence that amendments to a bill are more likely to solve the transfer problem as opposed to creating new transfer problems? We often hear about riders on a bill that promote some special interest unrelated to the main bill, and these riders are often attached via amendment.
It's not clear to me what the difference would be between proposing an amendment in futarchy and just proposing a new bill. So your special interest group proposes something that gains +16 for them, -20 for others, and bundles it with something that gives +5 to everybody. If someone notices this, why don't they just, immediately afterwards, propose a bill that reverses the +16 -20 part of the first bill? That proposal would be +4 on net and would pass.
In the systems I consider in the post, amendments are easier to propose, but also pay less of a reward to the proposer.
I don't see where you said this. You said amendments could be proposed if the proposer pays a set fee, and you said bills could be proposed if the proposer wins an auction. Which is easier to propose depends on whether the set fee is higher or lower than the auction price, which you didn't say.
But let's say the set fee is lower. Then, how is an amendment different from just a cheaper bill? Why not propose an amendment that tosses out the entire text of the bill and replaces it with a new one?
I also don't see where you said anything about a reward being paid to the proposer, other than the indirect (presumably illicit) reward if the proposer's policy proposal includes transfers to the proposer. Of course the proposer also stands to gain if they bet on their policy improving total utility and are right, but I don't see how that's any different between a bill and an amendment either.
Here is where I spoke of a reward: "The winner submits a bill, pays their auction price, and will be paid some share of the market estimated value gain if futarchy approves some variation on their bill"
Amendments are more limited in scope, cheaper to propose, and get paid lower rewards.
Limited in scope how and by whom? Whether a part of a bill is a transfer that does not increase societal welfare is a very difficult thing to objectively evaluate. A judge determining this has to determine what does and does not increase societal welfare, which is what the prediction market is supposed to do.
Lost me on the ineffectiveness of medicine. I’d be dead back in the 50s and pretty lousy around 2000. We’re pretty lucky.
I think you meant to reply to the main post.
As your penultimate paragraph indicates, relying less on plain text for policy contents and rather on a well-structured parameter space opens up mathematical analysis and search for optimality. Basically, policy-as-code.
A perhaps less ambitious but more immediately amenable solution is to impose a system-level tax on transfers complexity: the more transfers, the higher the tax to be paid by the proposer (this is the most naive version). Similarly to eg what the ethereum virtual machine imposes through gas costs.
Why do we assume that there’s a tradeoff between redistribution (pie share transfers) and growth (changing the size of the pie)? I understand why this theoretically makes under many orthodox economic theories but I’m not convinced it’s empirically true. At least in American economic history, it seems like the highest levels of growth happened in the post-war era where there was significant government intervention in the economy and redistribution via taxes and unions, etc.
I understand that this correlation doesn’t imply causation, but it seems like the onus should be on free marketers to prove why their system is more efficient, not the other way around.
I guess you weren't persuaded by my concrete example.
I think you're misunderstanding the transfer problem. The transfer problem is not just "any time the govt transfers wealth from person A to person B." It's specifically when the govt does this in a way that does not increase the total utility. It can be very good for total utility to redistribute wealth from the richest 1% to the people, because a dollar makes a much bigger improvement to the life of a poor person than a rich person. That's not the transfer problem.
Really it can be the transfer problem when the govt reverses this - tax cuts are the transfer problem. They benefit a special interest group (the rich) at the cost of total utility.
Most taxes cause deadweight loss. Pigovian taxes are an exception which increase utility, but those are considered "regressive" precisely because they aren't indexed to income. In order to efficiently raise lots of revenue, you need a broad base of taxation, such as via VAT, which again is regarded as more regressive than most federal taxes in the US.
The marginal value of a dollar is much higher for a poor person than for a rich person. If we consider goods assignment, it is less efficient for a billionaire to have a 5th house costing $10 million, than for a thousand poor, struggling people to have $10,000 worth of staple goods each; each individual poor person will each experience a far greater change in happiness than the billionaire will, and there are a thousand poor people to only one billionaire. This is not a small effect.
Deadweight loss is in units of (Price * Quantity). But societal welfare is in units of (Happiness * Quantity). The happiness caused by a dollar worth of personal goods is strongly dependent on how much money that person already has. So it is beneficial to accept a large amount of deadweight loss, if doing so results in production shifting towards goods that benefit poorer people instead of richer people.
"Happiness" is not a unit. You don't have actual measurements of it. And the distribution of goods is not random, rather diminishing marginal utility affects the actual decisions people make in acquiring them. We can infer its existence from such decisions. It is to the "revealed preference" of decisions we can look to rather than theorizing a priori about people's utility functions.
Also, "revealed preference" of purchasing decisions is weighted by money. For a poor person, spending $10,000 reveals a very strong preference or need for something. For Jeff Bezos, it reveals nothing more than mild interest; that amount of money is nothing to him. The amount of money a person spends on things is not an accurate reflection of how much those things are preferred, unless you somehow correct for the amount of money the person has.
Happiness can be measured (somewhat inaccurately) with polls, and of course we have a lot of common sense understanding of it.
Revealed preference comes not just in the form of purchasing decisions, but also in what one will do to acquire money.
How do polls "measure" happiness? In what units?
It's hard to measure happiness but no one doubts it exists. An ideal society should be maximizing some weighted metric of happiness, health, education, scientific advance, security.
Also, deadweight loss applies only when the government is taxing a business for each unit of goods produced. If we are instead speaking of a personal income or wealth tax, the deadweight loss diagram does not apply, because an individual's decisions about how hard to work are not strongly dependent on how much they are paid. A billionaire does not work a thousand times harder than a millionaire. Beyond a certain point, money and the luxury goods bought with it are just a way of keeping score, and people care more about making more money than their peer group than they do about having any particular absolute amount. If a peer group of rich people each have their income cut in half, they would still act in much the same way - same as if you cut the number of points scored by each basket in a basketball game in half, the players still do the same things.
(Rather unthinkable to cut a group of rich people's income in half - but what are the numbers on how many times the richest people's incomes have multiplied, inflation-adjusted, since the 80s? It's a lot more than 2. And yet their behavior hasn't changed much.)
Education is not an end, but a means. And when it largely consists in signalling, it collectively achieves little.
> If we are instead speaking of a personal income or wealth tax, the deadweight loss diagram does not apply, because an individual's decisions about how hard to work are not strongly dependent on how much they are paid.
You don't know what you're talking about. The decade with the greatest increase of women into the labor force was the 1980s, because of cuts to marginal tax rates. Women were marginal workers. People make decisions about entering or leaving the labor force (including how long they stay in school or when to retire), how many hours to work if they are part time, which jobs to accept when they are looking for work, etc.
> A billionaire does not work a thousand times harder than a millionaire.
In what units are you measuring how "hard" someone works?
> If a peer group of rich people each have their income cut in half, they would still act in much the same way - same as if you cut the number of points scored by each basket in a basketball game in half, the players still do the same things.
You make assertions, and yet economists have studied this question and come to different conclusions. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1674785
I don't think you have any measurements of the behavior of the "richest people", but instead just more assertions.
LOL if there really was a patriarchy, women would fight wars, and men would be exempt. Ever thought about that?
You seem to be off-topic. Did you intend to comment on a different post?
AI account?
This blog post is not about AI. And if you're suggesting I'm AI, note that I've been commenting on OB since it started in the '00s. I've also been blogging at https://entitledtoanopinion.wordpress.com/ since 2007.
Of course, I was speculating that 'Frank' was perhaps AI or AI driven and meant to post somewhere else but got mixed up somehow. Wasn't accusing you, sorry.