4th Grade Nation State

[Camicon]

Can't ignore something that was never brought to my attention.

In any case, that paper operates under the assumption that intensity of belief has as much a place in driving public policy as does the prevalence of said belief among the population, on the basis that the "socially optimal outcome" includes both. Discarding that ridiculousness renders the entire rest of the paper moot.

That's not really how you should be thinking about it. For example, if there's a policy which will cost one person £1,000 if implemented, but save 2 people £1 each. By majority voting, that policy would pass. By their bidding mechanism, the person who would lose £1,000 could compensate the other two in exchange for voting against the policy. All three of them end up better off. You're right that preference intensity is an arbitrary way of making group decisions, but so is preference prevalence to an extent. You can always find a set of circumstances in which either one is sub-optimal.

Preference prevalence is how democracies operate, from pure direct to FPTP representative. People are not elected based on how strongly voters feel, and legislation is not passed based on how strongly MP's feel. The only part which intensity of belief plays is by increasing the prevalence of belief, as it should be.

[The Joseon Dynasty]

That's not really how you should be thinking about it. For example, if there's a policy which will cost one person £1,000 if implemented, but save 2 people £1 each. By majority voting, that policy would pass. By their bidding mechanism, the person who would lose £1,000 could compensate the other two in exchange for voting against the policy. All three of them end up better off. You're right that preference intensity is an arbitrary way of making group decisions, but so is preference prevalence to an extent. You can always find a set of circumstances in which either one is sub-optimal.

Preference prevalence is how democracies operate, from pure direct to FPTP representative. People are not elected based on how strongly voters feel, and legislation is not passed based on how strongly MP's feel. The only part which intensity of belief plays is by increasing the prevalence of belief, as it should be.

I'm aware of how majority voting operates. That paper and the surrounding literature have been developed to provide alternatives for circumstances when majority voting might not be the most appropriate decision mechanism; for instance, when a policy is very harmful to a small group but slightly beneficial to a much larger group -- the textbook definition of tyranny of the majority. Furthermore, preference intensity doesn't necessarily have anything to do with preference prevalence. If, for example, preferences over a policy are defined by something fixed like racial identity, then you're not going to see much of a relationship between how strongly people feel and how many people feel that way.

I disagree with Xero about how his model works, but he's not incorrect that taking into account the strength of people's preferences can be a good way to make group decisions.

[Salandriagado]

Preference prevalence is how democracies operate, from pure direct to FPTP representative. People are not elected based on how strongly voters feel, and legislation is not passed based on how strongly MP's feel. The only part which intensity of belief plays is by increasing the prevalence of belief, as it should be.

I'm aware of how majority voting operates. That paper and the surrounding literature have been developed to provide alternatives for circumstances when majority voting might not be the most appropriate decision mechanism; for instance, when a policy is very harmful to a small group but slightly beneficial to a much larger group -- the textbook definition of tyranny of the majority. Furthermore, preference intensity doesn't necessarily have anything to do with preference prevalence. If, for example, preferences over a policy are defined by something fixed like racial identity, then you're not going to see much of a relationship between how strongly people feel and how many people feel that way.

I disagree with Xero about how his model works, but he's not incorrect that taking into account the strength of people's preferences can be a good way to make group decisions.

But we have actually sensible ideas for that: for example, ranked preference voting.

[Xerographica]

That isn't my argument. My argument is that the unequal value people place on money means it will always be a fundamentally flawed metric to measure how much people care about things. It's fine for examining what issues Hassan values over others, but it can never reliably measure the degrees of difference between how much Hassan and Jasmeen care about the same issues.

Your proposed ideology operates under the assumption that one dollar from a person with no disposable income carries the same weight as one dollar from a person with billions of dollars of disposable income. I trust you can see why that would pose a problem to your "coasianism"?

Let's say that I have a lemon tree that's full of lemons but I don't have any oranges. You, on the other hand, have an orange tree that's full of oranges but you don't have any lemons. I want oranges and you want lemons. It's a double coincidence of wants. Therefore? Therefore we trade.

Let's say that I have a lemon tree that's full of lemons but I don't have any money. You, on the other hand, have a ton of money but you don't have any lemons. I want money and you want lemons... therefore? Therefore we trade.

Except, you're arguing that we shouldn't be allowed to trade because I value money a lot more than you do. Eh? I value money a lot more than you do... therefore you shouldn't be allowed to give me your money?

If I have more lemons than money... and you have more money than lemons... then why would you argue against us trading? If we traded then we'd both be better off. We'd both have lemons and money.

Coasianism is trade. So if you're saying that coasianism is a bad idea because some people have more money than other people... then you're saying that trade is a bad idea because some people have more money than other people.

[Xerographica]

I disagree with Xero about how his model works, but he's not incorrect that taking into account the strength of people's preferences can be a good way to make group decisions.

Would my model improve if it was based on quadratic voting (QV)? If so, why?

Right now Michelle's class is using my model. What if we added QV to my model?

Last spring, however, my colleague Glen Weyl mentioned an idea along these lines that was so simple and elegant that I was amazed no one had ever thought of it before. In Glen’s voting mechanism, every voter can vote as many times as he or she likes. The catch, however, is that you have to pay each time you vote, and the amount you have to pay is a function of the square of the number of votes you cast. As a consequence, each extra vote you cast costs more than the previous vote. Just for the sake of argument, let’s say the first vote costs you $1. Then to vote a second time would cost $4. The third vote would be $9, the fourth $16, and so on. One hundred votes would cost you $10,000. So eventually, no matter how much you like a candidate, you choose to vote a finite number of times. - Steven D. Levitt, An Alternative to Democracy?

Let's say that Jacob wants to buy 4 votes. He would have to pay the teacher 16 pennies (4 x 4). Jacob casts his 4 votes for his preferred flag. Whichever flag gets the most votes is chosen as the class flag. All the money that is spent purchasing votes is given to the losers.

The stronger a student's preference for a flag, the more votes that they will be willing to purchase. But this is the same exact thing as saying.... the stronger a student's preference for a flag, the more money that they will be willing to pay.

QV model: willingness to pay for votes
My model: willingness to pay

Why does forcing the kids to purchase votes improve the efficiency of the outcome? In both cases the efficiency of the outcome is based on the kids' willingness to pay.

My guess is that Weyl was trying to improve voting by adding spending. With my model, I replace voting with spending.

[Soldati Senza Confini]

That isn't my argument. My argument is that the unequal value people place on money means it will always be a fundamentally flawed metric to measure how much people care about things. It's fine for examining what issues Hassan values over others, but it can never reliably measure the degrees of difference between how much Hassan and Jasmeen care about the same issues.

Your proposed ideology operates under the assumption that one dollar from a person with no disposable income carries the same weight as one dollar from a person with billions of dollars of disposable income. I trust you can see why that would pose a problem to your "coasianism"?

Let's say that I have a lemon tree that's full of lemons but I don't have any oranges. You, on the other hand, have an orange tree that's full of oranges but you don't have any lemons. I want oranges and you want lemons. It's a double coincidence of wants. Therefore? Therefore we trade.

Let's say that I have a lemon tree that's full of lemons but I don't have any money. You, on the other hand, have a ton of money but you don't have any lemons. I want money and you want lemons... therefore? Therefore we trade.

Except, you're arguing that we shouldn't be allowed to trade because I value money a lot more than you do. Eh? I value money a lot more than you do... therefore you shouldn't be allowed to give me your money?

If I have more lemons than money... and you have more money than lemons... then why would you argue against us trading? If we traded then we'd both be better off. We'd both have lemons and money.

Coasianism is trade. So if you're saying that coasianism is a bad idea because some people have more money than other people... then you're saying that trade is a bad idea because some people have more money than other people.

There is a difference between your system and trade though.

Trade doesn't require to gamble away your money.

[Equalaria]

This is very unethical. Kids this age are not only incredibly impressionable, but are more prone to anger and to lose social cohesion with one another in the contexts of simulating politics. It would be nice to see things like gender expression be more to the front of class life, as it is beneficial for children to understand the complexity of gender, insofar as it can be expressed in many unique ways. Starting the dialogue this early helps promote open mindedness and offset intolerance

[Camicon]

That isn't my argument. My argument is that the unequal value people place on money means it will always be a fundamentally flawed metric to measure how much people care about things. It's fine for examining what issues Hassan values over others, but it can never reliably measure the degrees of difference between how much Hassan and Jasmeen care about the same issues.

Your proposed ideology operates under the assumption that one dollar from a person with no disposable income carries the same weight as one dollar from a person with billions of dollars of disposable income. I trust you can see why that would pose a problem to your "coasianism"?

Let's say that I have a lemon tree that's full of lemons but I don't have any oranges. You, on the other hand, have an orange tree that's full of oranges but you don't have any lemons. I want oranges and you want lemons. It's a double coincidence of wants. Therefore? Therefore we trade.

Let's say that I have a lemon tree that's full of lemons but I don't have any money. You, on the other hand, have a ton of money but you don't have any lemons. I want money and you want lemons... therefore? Therefore we trade.

Except, you're arguing that we shouldn't be allowed to trade because I value money a lot more than you do. Eh? I value money a lot more than you do... therefore you shouldn't be allowed to give me your money?

If I have more lemons than money... and you have more money than lemons... then why would you argue against us trading? If we traded then we'd both be better off. We'd both have lemons and money.

That is not what your system is.

Coasianism is trade. So if you're saying that coasianism is a bad idea because some people have more money than other people... then you're saying that trade is a bad idea because some people have more money than other people.

I'm saying that coasianism is a bad idea because you are using money as a metric to measure how much a person values things compared to other people. Because people value money differently this kind of comparison will never be accurate or reliable.

Preference prevalence is how democracies operate, from pure direct to FPTP representative. People are not elected based on how strongly voters feel, and legislation is not passed based on how strongly MP's feel. The only part which intensity of belief plays is by increasing the prevalence of belief, as it should be.

I'm aware of how majority voting operates. That paper and the surrounding literature have been developed to provide alternatives for circumstances when majority voting might not be the most appropriate decision mechanism; for instance, when a policy is very harmful to a small group but slightly beneficial to a much larger group -- the textbook definition of tyranny of the majority. Furthermore, preference intensity doesn't necessarily have anything to do with preference prevalence. If, for example, preferences over a policy are defined by something fixed like racial identity, then you're not going to see much of a relationship between how strongly people feel and how many people feel that way.

I disagree with Xero about how his model works, but he's not incorrect that taking into account the strength of people's preferences can be a good way to make group decisions.

If your system of government has a mechanism by which it measures intensity of belief, and that is given equal weight to the individual number of people who hold said belief, then you open up the door to a tyranny of the minority, the exact thing democracy is meant to stop (which FPTP does terribly, but that's neither here nor there).

This is very unethical.

No it isn't.

Kids this age are not only incredibly impressionable, but are more prone to anger and to lose social cohesion with one another in the contexts of simulating politics.

No they aren't.

It would be nice to see things like gender expression be more to the front of class life, as it is beneficial for children to understand the complexity of gender, insofar as it can be expressed in many unique ways. Starting the dialogue this early helps promote open mindedness and offset intolerance

Not sure what that has to do with the thread, but I agree.

[The Joseon Dynasty]

I'm aware of how majority voting operates. That paper and the surrounding literature have been developed to provide alternatives for circumstances when majority voting might not be the most appropriate decision mechanism; for instance, when a policy is very harmful to a small group but slightly beneficial to a much larger group -- the textbook definition of tyranny of the majority. Furthermore, preference intensity doesn't necessarily have anything to do with preference prevalence. If, for example, preferences over a policy are defined by something fixed like racial identity, then you're not going to see much of a relationship between how strongly people feel and how many people feel that way.

I disagree with Xero about how his model works, but he's not incorrect that taking into account the strength of people's preferences can be a good way to make group decisions.

But we have actually sensible ideas for that: for example, ranked preference voting.

They solve different problems. Ranked preference voting is on an ordinal scale; it doesn't take into account the distance between your preference for each option. If you apply it to the scenario I gave a couple of posts ago (i.e. a policy such that one person loses £1,000, two people gain £1), any ranked preference voting system is just going to reduce to a majority voting system.

I'm aware of how majority voting operates. That paper and the surrounding literature have been developed to provide alternatives for circumstances when majority voting might not be the most appropriate decision mechanism; for instance, when a policy is very harmful to a small group but slightly beneficial to a much larger group -- the textbook definition of tyranny of the majority. Furthermore, preference intensity doesn't necessarily have anything to do with preference prevalence. If, for example, preferences over a policy are defined by something fixed like racial identity, then you're not going to see much of a relationship between how strongly people feel and how many people feel that way.

I disagree with Xero about how his model works, but he's not incorrect that taking into account the strength of people's preferences can be a good way to make group decisions.

If your system of government has a mechanism by which it measures intensity of belief, and that is given equal weight to the individual number of people who hold said belief, then you open up the door to a tyranny of the minority, the exact thing democracy is meant to stop (which FPTP does terribly, but that's neither here nor there).

Although Xero thinks otherwise, the point of a decision mechanism like the one in that paper isn't to apply it to absolutely everything. It's an alternative for specific circumstances where majority voting, or any ordinal voting system, is inappropriate.

With that said, you're also misdiagnosing the problem with it. A minority group is only going to have a disproportional influence over the outcome of the bid if the majority group is apathetic or just cares about the issue less. This is true in majority voting as well; if people don't turn up for elections, a smaller segment of the population is going to decide the outcome. It's no different here.

I disagree with Xero about how his model works, but he's not incorrect that taking into account the strength of people's preferences can be a good way to make group decisions.

Would my model improve if it was based on quadratic voting (QV)? If so, why?

Right now Michelle's class is using my model. What if we added QV to my model?

Last spring, however, my colleague Glen Weyl mentioned an idea along these lines that was so simple and elegant that I was amazed no one had ever thought of it before. In Glen’s voting mechanism, every voter can vote as many times as he or she likes. The catch, however, is that you have to pay each time you vote, and the amount you have to pay is a function of the square of the number of votes you cast. As a consequence, each extra vote you cast costs more than the previous vote. Just for the sake of argument, let’s say the first vote costs you $1. Then to vote a second time would cost $4. The third vote would be $9, the fourth $16, and so on. One hundred votes would cost you $10,000. So eventually, no matter how much you like a candidate, you choose to vote a finite number of times. - Steven D. Levitt, An Alternative to Democracy?

Let's say that Jacob wants to buy 4 votes. He would have to pay the teacher 16 pennies (4 x 4). Jacob casts his 4 votes for his preferred flag. Whichever flag gets the most votes is chosen as the class flag. All the money that is spent purchasing votes is given to the losers.

The stronger a student's preference for a flag, the more votes that they will be willing to purchase. But this is the same exact thing as saying.... the stronger a student's preference for a flag, the more money that they will be willing to pay.

QV model: willingness to pay for votes
My model: willingness to pay

Why does forcing the kids to purchase votes improve the efficiency of the outcome? In both cases the efficiency of the outcome is based on the kids' willingness to pay.

My guess is that Weyl was trying to improve voting by adding spending. With my model, I replace voting with spending.

Look at Proposition 2 in the paper you linked. The expression v_i (2G(b_i) - 1) is voter i's expected utility given a bid b_i. G(b_i) is a symmetric distribution function. If you take a first order Taylor expansion of G(b_i) around 0, then you have that voter i's expected utility is a product of v_i and b_i; i.e. it is linear in b_i and the marginal benefit is v_i. Since we want it to be the case that voters bid as their true valuation, we must choose a cost function so that the optimal bid occurs when b_i = v_i. This will only occur when the cost function is quadratic, since a quadratic cost function has the property that the marginal cost is equal to how many bids you already have. That's the reasoning behind it.

I'm sorry it can't be more intuitive. It really is something that comes right out of the model itself. The takeaway is that if you have, for example, a linear cost function like yours, it's possible that voters will bid well past or well below their true valuation.

Edit: Made a couple of corrections.

[Galloism]

Look at Proposition 2 in the paper you linked. The expression v_i (2G(b_i) - 1) is voter i's expected utility given a bid b_i. G(b_i) is a symmetric distribution function. If you take a first order Taylor expansion of G(b_i) around 0, then you have that voter i's expected utility is a product of v_i and b_i. If voting at your true valuation is optimal, then this means that voter i's expected utility must be quadratic in v_i. Since we want it to be the case that voters bid as their true valuation, we must choose a cost function so that marginal cost equals marginal benefit at the true valuation. This will only occur when the cost function is also quadratic. That's the reasoning behind it.

I'm sorry it can't be more intuitive. It really is something that comes right out of the model itself. The takeaway is that if you have, for example, a linear cost function like yours, it's possible that voters will bid well past their true valuation.

I'll be honest - that's well above my head, other than the model is designed in such a way so that people actually bid (close to) their true valuation, because that's actually the optimal amount to bid mathematically speaking.

[The Joseon Dynasty]

Look at Proposition 2 in the paper you linked. The expression v_i (2G(b_i) - 1) is voter i's expected utility given a bid b_i. G(b_i) is a symmetric distribution function. If you take a first order Taylor expansion of G(b_i) around 0, then you have that voter i's expected utility is a product of v_i and b_i. If voting at your true valuation is optimal, then this means that voter i's expected utility must be quadratic in v_i. Since we want it to be the case that voters bid as their true valuation, we must choose a cost function so that marginal cost equals marginal benefit at the true valuation. This will only occur when the cost function is also quadratic. That's the reasoning behind it.

I'm sorry it can't be more intuitive. It really is something that comes right out of the model itself. The takeaway is that if you have, for example, a linear cost function like yours, it's possible that voters will bid well past their true valuation.

I'll be honest - that's well above my head, other than the model is designed in such a way so that people actually bid (close to) their true valuation, because that's actually the optimal amount to bid mathematically speaking.

Yeah, that's pretty much it. I guess the intuition is that, with a quadratic cost function, the cost of an additional bid is equal to how many bids I have already bought. The benefit of an additional bid is equal to my true valuation of winning. So with a quadratic cost function, I will want to continue buying bids until the cost of an additional bid is the same as the benefit of an additional bid, which is exactly my true valuation. A linear cost function means that the cost of an additional bid is fixed. If that cost is below my true valuation, I will want to buy bids indefinitely. If it's above my valuation, I won't want to buy any bids.

Edit: Didn't like my previous explanation...

[Salandriagado]

But we have actually sensible ideas for that: for example, ranked preference voting.

They solve different problems. Ranked preference voting is on an ordinal scale; it doesn't take into account the distance between your preference for each option. If you apply it to the scenario I gave a couple of posts ago (i.e. a policy such that one person loses £1,000, two people gain £1), any ranked preference voting system is just going to reduce to a majority voting system.

Then use a different sensible system: for example, if you have a large number of votes to make (which most situations will over time), you give each person (say) ten votes per issue, but let the unused votes carry over to the next motion. Not sure if it enjoys the perfect representation property of the system in the paper (in fact, I'm pretty sure it doesn't), and it definitely has its problems, but it should certainly be better than Xero's system in terms of measuring actual preferences, given sensible limits with regards to issue spamming and such, since it removes the differential supply problem and removes the exponential growth in voting power from repeatedly "losing" votes.

[Camicon]

If your system of government has a mechanism by which it measures intensity of belief, and that is given equal weight to the individual number of people who hold said belief, then you open up the door to a tyranny of the minority, the exact thing democracy is meant to stop (which FPTP does terribly, but that's neither here nor there).

Although Xero thinks otherwise, the point of a decision mechanism like the one in that paper isn't to apply it to absolutely everything. It's an alternative for specific circumstances where majority voting, or any ordinal voting system, is inappropriate.

With that said, you're also misdiagnosing the problem with it. A minority group is only going to have a disproportional influence over the outcome of the bid if the majority group is apathetic or just cares about the issue less. This is true in majority voting as well; if people don't turn up for elections, a smaller segment of the population is going to decide the outcome. It's no different here.

I'd prefer the occasional "inefficient" majority vote than the spectre of a "tyranny of the minority" situation.

In any case, not voting - implicit approval of a smaller segment of the population having disproportionate influence - is not the same as giving electoral value to opinions.

Let's say slightly over half the population of Country X doesn't cast a ballot in their federal election. Whatever their reasoning, they have abrogated their right to influence the outcome of the election to their fellow citizens. Their lack of participation is voluntary, and so even though whatever government that leads Country X after the election will have attained power with less than half the country voting for them - even if every voters casts for them - this isn't a true "tyranny of the minority" situations. By choosing not to vote a majority of the electorate has effectively seceded their opinion to the minority. Unless Country X uses a FPTP system, which blows just so hard, you won't find a situation where a minority of voters are ruling over a majority of voters - I think. I can't recall if there are representative systems other than FPTP that allow a plurality or minority to form a majority government, please correct me if I am wrong.

Now, let's say that Country X has mandatory voting; everyone of voting age casts a ballot, but they also assign electoral value to how strongly they feeling about the ballot they are casting. This creates the very real possibility that a majority government could be formed despite receiving a minority of votes cast. You could see as a 34/66 split, provided the 34% are determined to feel twice as strongly about their vote than the 66%. Even if you put in safeguards to avoid such obviously lopsided outcomes there could still be a situation where a minority will have full control over the majority. This is not part of the country voluntarily seceding their opinion to the minority, this is a minority of the country taking majority control because they feel more strongly about it than their opponents; a "I AM SHOUTING MORE LOUDLY THAN YOU SO THAT MEANS I AM IN CHARGE" situation. And I am not comfortable with a system that allows that, whether it be used in a federal election or in a local zoning dispute.

(I'm sure you understand all of this already, but I felt it prudent to be as clear as possible about my own thinking)

Xero's proposed system is a few steps shittier, by equating the spending of money with intensity of belief, but that just adds one more broken component to an already fundamentally wrecked foundation.

[The Joseon Dynasty]

They solve different problems. Ranked preference voting is on an ordinal scale; it doesn't take into account the distance between your preference for each option. If you apply it to the scenario I gave a couple of posts ago (i.e. a policy such that one person loses £1,000, two people gain £1), any ranked preference voting system is just going to reduce to a majority voting system.

Then use a different sensible system: for example, if you have a large number of votes to make (which most situations will over time), you give each person (say) ten votes per issue, but let the unused votes carry over to the next motion. Not sure if it enjoys the perfect representation property of the system in the paper (in fact, I'm pretty sure it doesn't), and it definitely has its problems, but it should certainly be better than Xero's system in terms of measuring actual preferences, given sensible limits with regards to issue spamming and such, since it removes the differential supply problem and removes the exponential growth in voting power from repeatedly "losing" votes.

Yeah, I think we're all agreed that Xero's system is rubbish, and you could probably come up with countless alternatives that are better than it. I was just disputing the idea that the paper he linked to, which is written by actual researchers who know what they're doing, is by extension also rubbish. That paper isn't very practical in terms of how you'd actually implement such a voting system, but, as you said, it has the very nice property of incentivising voters to reveal their true preferences, which is an incredibly difficult property to achieve.

[Xerographica]

John Oliver seems to have read the NY Times article on hedging that I shared earlier in this thread.

In the Nude Beach Economics thread, I shared this quote...

If a pretty woman strolls through the hotel lobby many tired convention delegates may get some external benefits, but, presumably, she finds it to her own advantage to stroll, and few delegates would pay her to stroll more than she already does. - James Buchanan, What Should Economists Do?

Today I saw this...

Besides, why would you want to wear something that's seen on dozens of other women every day, everywhere? I thought women didn't like doing that for obvious reasons. Yoga pants belong in the yoga studio. What's next? Wearing a "Speedo" to the supermarket? Imagine if men did that. Yuck!

To all yoga pant wearers, I struggle with my own physicality as I age. I don't want to struggle with yours. Thanks - Alan Sorrentino, Letter: Please, women, put away the yoga pants

Of course lots of women were offended. In fact, many of them paraded past his house wearing their yoga pants.

What Camicon doesn't seem to understand/appreciate/grasp is the concept of negative/positive externalities. I derive a certain amount of enjoyment/benefit/utility from drinking alcohol in one year. If enough voters wanted to, they could take this benefit away from me. Our history proves that this possibility is not far-fetched. The majority's decision would harm me. Their choice would have a negative spillover (my harm). It's the simple concept of tyranny of the majority. Clearly tyranny is harmful.

But if voters are going to take away my right to drink alcohol... then why shouldn't they compensate me for my loss of benefit? Their benefit should be greater than my cost/harm. Because if it isn't, then it doesn't make sense for alcohol to be made illegal.

In order to ensure the most beneficial outcome, we would simply replace voting with spending.

The Joseon Dynasty understands the concept of externalities. However, he feels that the quadratic voting (QV) model is superior to my model. I asked him to clarify his position. Unfortunately, his explanation really did not clear things up.

Using the example of alcohol... if we are going to use coasianism to decide whether it should be illegal.... then I'd have to decide how much benefit I derive from alcohol in one year. Let's say that I decide I derive $200 dollars worth of benefit from drinking alcohol. If alcohol was made illegal, then this $200 dollars of benefit would be taken away from me.

Now, according to The Joseon Dynasty, somehow my valuation is wrong. He thinks that I can only accurately figure out how much I value alcohol for one year if I'm given the opportunity to quadratically buy votes. This means that I'd spend more, or less, than $200 dollars on votes. Eh? What? Why would I spend more, or less, money on votes than my valuation? How does that make any sense?

I decide that my valuation is $200 dollars... but then I decide to buy $100 dollars worth of votes. What happened? Why the disparity? Did I change my mind? Was I lying to myself when I decided that my valuation is $200 dollars? Did the act of actually buying votes force me to be honest?

Now, imagine the implications if it was true that buying votes actually does improve the accuracy of our valuations. Rather than simply buy a $17,000 dollar car, you'd buy votes for that car. If you only bought $12,000 dollars worth of votes... then obviously you shouldn't buy a $17,000 dollar car. You should buy a $12,000 dollar car. Because it turns out that your true valuation of a car isn't $17,000 dollars... it's $12,000 dollars. Clearly you would want to use this same approach before you bought a computer or a house.

Maybe I'm missing something. But if I say that my true valuation of drinking alcohol for one year is $200 dollars... it's rather absurd for The Joseon Dynasty, or anybody else, to tell me that my valuation is wrong. How could he possibly know how much benefit I derive from drinking alcohol for a year? How could he possible know how much I'd be harmed if it was illegal for me to drink alcohol for a year?

Today Sergio purchased this quote for 51 cents...

It is thus that the private interests and passions of individuals naturally dispose them to turn their stocks towards the employments which in ordinary cases are most advantageous to the society. But if from this natural preference they should turn too much of it towards those employments, the fall of profit in them and the rise of it in all others immediately dispose them to alter this faulty distribution. Without any intervention of law, therefore, the private interests and passions of men naturally lead them to divide and distribute the stock of every society among all the different employments carried on in it as nearly as possible in the proportion which is most agreeable to the interest of the whole society. - Adam Smith, Wealth of Nations

This is the Invisible Hand. Our valuations/allocations/purchases/payments/sacrifices alter each other's behavior. The more accurate our valuations, the more beneficial our behavior.

Before, during or after the auction... Jacob asked if he could auction his quote. He might have mentioned, or maybe Michelle asked... but it turned out that he had copied his quote... but he had left the copy at home. Michelle said that he could do whatever he wanted with his quote but maybe he might want to try and sell his copy instead.

Not sure how long afterwards, Sergio decided to sell his quote to Elizabeth for 41 cents. He sold her his original and didn't think to make a copy first. It was the first time, as far as we know, that one student sold their quote to another student. Michelle took a photo and texted it to me.

Michelle auctioned another quote today...

In San Francisco, a new form of covert agriculture is taking root: making fruit trees out of trees that normally don’t bear fruit, and turning an entire city into an orchard. - Andrew Price, Guerrilla Grafters: Splicing Fruit-Bearing Branches Onto City Trees

Laila won the quote for 41 cents. Maybe they'll turn their school into an orchard.

Just in case you missed it, when Michelle auctions the quotes, the kids who win them get to decide which class/school goods their money gets spent on. The departments haven't been set up yet but Michelle's keeping track of the winning bids. We were wondering whether the Gardening Department or the Book Department would receive more taxes. Most of the kids love to garden at the school... but lots of the kids also love books. Is the demand for books greater than the demand for plants?

My classroom experiment has been criticized for lack of formality. But informal experiments can answer questions. How long would it take for one student to sell their quote to another student? Turns out that it took less than a week. Is the demand for books greater than the demand for plants? Is this a good question? Yeah. Do we know the answer? Nope. Should we know the answer? Definitely.

[Camicon]

What Camicon doesn't seem to understand/appreciate/grasp is the concept of negative/positive externalities. I derive a certain amount of enjoyment/benefit/utility from drinking alcohol in one year. If enough voters wanted to, they could take this benefit away from me. Our history proves that this possibility is not far-fetched. The majority's decision would harm me. Their choice would have a negative spillover (my harm). It's the simple concept of tyranny of the majority. Clearly tyranny is harmful.

But if voters are going to take away my right to drink alcohol... then why shouldn't they compensate me for my loss of benefit? Their benefit should be greater than my cost/harm. Because if it isn't, then it doesn't make sense for alcohol to be made illegal.

In order to ensure the most beneficial outcome, we would simply replace voting with spending.

Maybe, instead of assuming I don't understand what the tyranny of the majority entails and adopting a patronizing tone, you could ask me.

Tyranny of any kind is harmful, yeah, but by definition a tyranny of the minority is more harmful than a tyranny of the majority. If forced to choose I will favour a system which allows for a tyranny of the majority over one which allows for a tyranny of the majority and of the minority.

That's why your system is shit, by the way. Not the "minority gets compensation" bit, but the fact that its design encourages a minority of extremely wealthy people to game the system to the detriment of an overwhelming majority.

[Xerographica]

What Camicon doesn't seem to understand/appreciate/grasp is the concept of negative/positive externalities. I derive a certain amount of enjoyment/benefit/utility from drinking alcohol in one year. If enough voters wanted to, they could take this benefit away from me. Our history proves that this possibility is not far-fetched. The majority's decision would harm me. Their choice would have a negative spillover (my harm). It's the simple concept of tyranny of the majority. Clearly tyranny is harmful.

But if voters are going to take away my right to drink alcohol... then why shouldn't they compensate me for my loss of benefit? Their benefit should be greater than my cost/harm. Because if it isn't, then it doesn't make sense for alcohol to be made illegal.

In order to ensure the most beneficial outcome, we would simply replace voting with spending.

Maybe, instead of assuming I don't understand what the tyranny of the majority entails and adopting a patronizing tone, you could ask me.

Tyranny of any kind is harmful, yeah, but by definition a tyranny of the minority is more harmful than a tyranny of the majority. If forced to choose I will favour a system which allows for a tyranny of the majority over one which allows for a tyranny of the majority and of the minority.

That's why your system is shit, by the way. Not the "minority gets compensation" bit, but the fact that its design encourages a minority of extremely wealthy people to game the system to the detriment of an overwhelming majority.

It's not at all unreasonable to assume that the vast majority of people aren't familiar with the economic concept of positive/negative externalities.

You agree that we can reasonably define tyranny as harm. But if we define tyranny as harm... and my system eliminates harm... then why, exactly, would you refer to my system as tyranny of the minority? Where's the tyranny? Where's the harm? With my system... the compensation/benefit is greater than the harm...

compensation > harm

This effectively eliminates the harm. My harm for alcohol being illegal for a year is $200 dollars. If my side loses... then it's a given that the compensation will be greater than the harm...

$300 (compensation ) > $200 (harm)

The compensation eliminates the harm. I lost $200 dollars worth of benefit but gained $300 dollars worth of compensation. I'm better off as a result of the trade.

Please do us all a favor and provide a specific scenario to illustrate your argument. For example, I used the right to drink alcohol to illustrate my argument. If you don't want to use my alcohol scenario... then you're welcome to come up with your own scenario.

[The Two Jerseys]

Just in case you missed it, when Michelle auctions the quotes, the kids who win them get to decide which class/school goods their money gets spent on.

So not only are you not solving the free-rider "problem", you're making it even easier to be a free-rider...

[Galloism]

Please do us all a favor and provide a specific scenario to illustrate your argument. For example, I used the right to drink alcohol to illustrate my argument. If you don't want to use my alcohol scenario... then you're welcome to come up with your own scenario.

I'll bite.

I value alcohol at $5,000 per year. I have exactly $200 to my name, and no credit, and no way to get more before bidding closes.

Prohibition comes up for bid. I bid my entire $200, choosing to go without food until my next paycheck. Because of hedgers, I am compensated $200.04 when I lose. But I lost $5,000 per year of alcohol enjoyment.

$5,000 - $200.04 = $4,799.96 of harm

[Camicon]

Maybe, instead of assuming I don't understand what the tyranny of the majority entails and adopting a patronizing tone, you could ask me.

Tyranny of any kind is harmful, yeah, but by definition a tyranny of the minority is more harmful than a tyranny of the majority. If forced to choose I will favour a system which allows for a tyranny of the majority over one which allows for a tyranny of the majority and of the minority.

That's why your system is shit, by the way. Not the "minority gets compensation" bit, but the fact that its design encourages a minority of extremely wealthy people to game the system to the detriment of an overwhelming majority.

It's not at all unreasonable to assume that the vast majority of people aren't familiar with the economic concept of positive/negative externalities.

And yet a dick move, nonetheless.

You agree that we can reasonably define tyranny as harm. But if we define tyranny as harm... and my system eliminates harm... then why, exactly, would you refer to my system as tyranny of the minority? Where's the tyranny? Where's the harm? With my system... the compensation/benefit is greater than the harm...

Your system does not eliminate harm. Providing monetary compensation to people who have been harmed does not mean they were not harmed. If Bill Gates kicks me in the balls and I get a million dollars in a settlement, that doesn't mean my balls weren't kicked. Compensation doesn't "eliminate" harm, it compensates for it.

compensation > harm

This effctively eliminates the harm. My harm for alcohol being illegal for a year is $200 dollars. If my side loses... then it's a given that the compensation will be greater than the harm...

$300 (compensation ) > $200 (harm)

The compensation eliminates the harm. I lost $200 dollars worth of benefit but gained $300 dollars worth of compensation. I'm better off as a result of the trade.

Please do us all a favor and provide a specific scenario to illustrate your argument. For example, I used the right to drink alcohol to illustrate my argument. If you don't want to use my alcohol scenario... then you're welcome to come up with your own scenario.

Our disagreement is baser than that, a scenario is not needed.


We have agreed that tyranny is harmful.
A tyranny of the minority sees a majority people living under tyranny; the reverse is true for a tyranny of the majority.
Ergo, tyranny of the minority is more harmful than tyranny of the majority.

Your system allows for, even encourages, tyranny of the minority - specifically of the ultra-wealthy.
Democracies, excluding representative democracies which utilize FPTP, do not allow for tyranny of the minority.
Because your system allows for and encourages tyranny of the minority, which most democracies do not, your system is more harmful.

[Xerographica]

Please do us all a favor and provide a specific scenario to illustrate your argument. For example, I used the right to drink alcohol to illustrate my argument. If you don't want to use my alcohol scenario... then you're welcome to come up with your own scenario.

I'll bite.

I value alcohol at $5,000 per year. I have exactly $200 to my name, and no credit, and no way to get more before bidding closes.

Prohibition comes up for bid. I bid my entire $200, choosing to go without food until my next paycheck. Because of hedgers, I am compensated $200.04 when I lose. But I lost $5,000 per year of alcohol enjoyment.

$5,000 - $200.04 = $4,799.96 of harm

You can't go up to a Lamborghini dealer and say that your valuation of one of his cars is a million dollars and expect him to just give it to you. If you want him to give you the car, you need to give him enough money. And when you give him the money, then he'll believe that your valuation is at least as large as your payment.

So you say that your valuation of alcohol is $5,000 dollars... but you only spend $200 dollars to oppose prohibition. Do I trust your words... or your actions? Clearly I'm going to trust your actions. Does this mean that you're lying? Not necessarily. The fact of the matter is that I'm not a mind reader. Nobody is a mind reader.

You can argue that coasianism doesn't work because your valuation is greater than your ability to pay. But this really wouldn't just be true of coasianism... this would be true of markets in general.

I can concede defeat, but only if you promise to put as much effort into eliminating markets as I've put into promoting them. If you don't uphold your end of the bargain, then clearly you don't genuinely believe that your argument has any real merit.

So do you genuinely believe that your argument has any real merit? If so, I'll concede defeat and you'll start a thread explaining why we should get rid of markets entirely. And you'll continue to do so for at least 4 years.

[Lost heros]

Xero, are you going to address the issue that via hedging the wealthy can have an absurdly high influence, but minimize payouts to everyone else? Are you going to address that via hedging the results do not resemble the actual value of the populaces? Are you going to address that for an equivalent WTP across the board everyone must have access to the same relative amount of wealth ($10 for a poor person is more valuable than $10 for a rich person)?

The whole basis of what you are proposing is to weigh in value with votes, but these three points show that your system does not do such.

With hedging by the wealthy, the wealthy can afford to bid HIGHER than their WTP, win the vote, and have LOW losses.
For example, let's say John, a multi-billionaire, Steve, and Jessica, who are average people, use your system to decide where to go to dinner. John wants to go to Chipotle, but Steve and Jessica want to go to Moe's. They all value their restaurant at $10. So Steve and Jessica bid $10 on Moe's. John bids $1,000,000,000 on Chipotle and $999,999,075 (The difference being just over double his adversaries WTP [double because 2 people are bidding against him]) on Moe's. Chipotle wins. Steve and Jessica get paid $10.0 each (1,000,000,000/999,999,095 * 10,) and John keeps the rest because he has more and thus more influence only has to pay $20 to win and can also ensure that the others will never get to his level.

Not only that, but because of hedging people will not bid their WTP. I know that the above scenario wasn't entirely realistic, but it shows something important. Because of hedging, John bets on the other side, where he is not willing to pay while on his side he bets A LOT more than what he is actually willing to pay, but the point remains the same. The TRUE WTP for the above decision was $10 (arguably $20) for Chipotle and $20 for Moe's, but that doesn't match up with the true bidded amount because John can win easily if he hedges absurdly high because he can and still pay out roughly around his true WTP.

Furthermore, assuming that everyone bids their WTP (presumably they are all idiots) each WTP do not have the same relative base. The fact of the matter is, a person who only has $100 in disposable income values $10 more than someone who has $1000 in disposable incomes, and if they each spent $10 on opposite sides it would have equal weight, but it wouldn't have equal VALUE.

Are you going to respond to this?

[The Joseon Dynasty]

In order to ensure the most beneficial outcome, we would simply replace voting with spending.

The Joseon Dynasty understands the concept of externalities. However, he feels that the quadratic voting (QV) model is superior to my model. I asked him to clarify his position. Unfortunately, his explanation really did not clear things up.

Using the example of alcohol... if we are going to use coasianism to decide whether it should be illegal.... then I'd have to decide how much benefit I derive from alcohol in one year. Let's say that I decide I derive $200 dollars worth of benefit from drinking alcohol. If alcohol was made illegal, then this $200 dollars of benefit would be taken away from me.

Now, according to The Joseon Dynasty, somehow my valuation is wrong. He thinks that I can only accurately figure out how much I value alcohol for one year if I'm given the opportunity to quadratically buy votes. This means that I'd spend more, or less, than $200 dollars on votes. Eh? What? Why would I spend more, or less, money on votes than my valuation? How does that make any sense?

I decide that my valuation is $200 dollars... but then I decide to buy $100 dollars worth of votes. What happened? Why the disparity? Did I change my mind? Was I lying to myself when I decided that my valuation is $200 dollars? Did the act of actually buying votes force me to be honest?

Now, imagine the implications if it was true that buying votes actually does improve the accuracy of our valuations. Rather than simply buy a $17,000 dollar car, you'd buy votes for that car. If you only bought $12,000 dollars worth of votes... then obviously you shouldn't buy a $17,000 dollar car. You should buy a $12,000 dollar car. Because it turns out that your true valuation of a car isn't $17,000 dollars... it's $12,000 dollars. Clearly you would want to use this same approach before you bought a computer or a house.

Maybe I'm missing something. But if I say that my true valuation of drinking alcohol for one year is $200 dollars... it's rather absurd for The Joseon Dynasty, or anybody else, to tell me that my valuation is wrong. How could he possibly know how much benefit I derive from drinking alcohol for a year? How could he possible know how much I'd be harmed if it was illegal for me to drink alcohol for a year?

Because you've simplified the problem too much. Let's say that I value alcohol by £200 and there's a vote about whether to implement prohibition. When deciding how much to bid for alcohol, I'm considering both how much I value alcohol (£200) and how likely it is that prohibition will pass. How likely it is that prohibition passes is determined by how much I bid, but also by how much everyone else bids. Assuming I get £0 if prohibition passes (i.e. no consumption of alcohol), then the objective I'm really trying to maximise is P(X)*£200, where P(X) is the probability that prohibition does not pass given that I bid £X against it. This is because I'm only going to consume my £200 of alcohol if prohibition does not pass.

Actually maximising this involves some maths that I won't get into (i.e. first order Taylor expansion). Just assume that P(X) = X/1000 or something. So what I'm trying to maximise with my bid X is £200 * (X / 1000). This means that each extra pound I bid is going to improve my expected outcome by £200 * (1 / 1000) = £0.2. Let's suppose, as you do, that each bid just costs me the amount that I'm bidding, so the cost is simply C(X) = -X for a bid of X.
Then my objective now looks like £200 * (X / 1000) - X. Since my benefit of an additional pound bid is £0.2 but that bid costs me £1, I will choose not to bid at all. Let's suppose the cost function is quadratic and scaled by the distribution (as they do in the paper), then the cost function is C(X) = X^2 / 2000. Then, supposing I've spent X so far, adding an additional £1 to my bid is going to get me £200 * (1 / 1000) - X / 1000. If I solve this for X, I get that X = £200. That is going to be the optimal amount I will bid.

So recap: Under your cost function, I will spend £0. Under my cost function, I will spend my true valuation £200.

That's the simplest possible explanation I can give you of that.

[Xerographica]

Xero, are you going to address the issue that via hedging the wealthy can have an absurdly high influence, but minimize payouts to everyone else? Are you going to address that via hedging the results do not resemble the actual value of the populaces? Are you going to address that for an equivalent WTP across the board everyone must have access to the same relative amount of wealth ($10 for a poor person is more valuable than $10 for a rich person)?

The whole basis of what you are proposing is to weigh in value with votes, but these three points show that your system does not do such.

With hedging by the wealthy, the wealthy can afford to bid HIGHER than their WTP, win the vote, and have LOW losses.
For example, let's say John, a multi-billionaire, Steve, and Jessica, who are average people, use your system to decide where to go to dinner. John wants to go to Chipotle, but Steve and Jessica want to go to Moe's. They all value their restaurant at $10. So Steve and Jessica bid $10 on Moe's. John bids $1,000,000,000 on Chipotle and $999,999,075 (The difference being just over double his adversaries WTP [double because 2 people are bidding against him]) on Moe's. Chipotle wins. Steve and Jessica get paid $10.0 each (1,000,000,000/999,999,095 * 10,) and John keeps the rest because he has more and thus more influence only has to pay $20 to win and can also ensure that the others will never get to his level.

Not only that, but because of hedging people will not bid their WTP. I know that the above scenario wasn't entirely realistic, but it shows something important. Because of hedging, John bets on the other side, where he is not willing to pay while on his side he bets A LOT more than what he is actually willing to pay, but the point remains the same. The TRUE WTP for the above decision was $10 (arguably $20) for Chipotle and $20 for Moe's, but that doesn't match up with the true bidded amount because John can win easily if he hedges absurdly high because he can and still pay out roughly around his true WTP.

Furthermore, assuming that everyone bids their WTP (presumably they are all idiots) each WTP do not have the same relative base. The fact of the matter is, a person who only has $100 in disposable income values $10 more than someone who has $1000 in disposable incomes, and if they each spent $10 on opposite sides it would have equal weight, but it wouldn't have equal VALUE.

Are you going to respond to this?

I've responded to all this throughout the thread. I feel reluctant to be redundant.

Galloism is planning on testing out the hedging issue. If you go back and actually read my posts, then I said something like once burned twice shy. Meaning... Samantha might hedge her bets on whether class is inside or outside. Let's say that the result of the trade is that class is outside and she gets $7 dollars. Was this a good deal for Samantha? The answer depends entirely on her valuation of having class outside. If her valuation of having class outside is -$10 dollars, then clearly it was a mistake for her to hedge her bets.

I think that enough of the students will learn that honesty is the best policy.

Regarding the fact that people don't all have the same amount of money... I explained why this is a non-issue to Camicon.

You have lots of lemons but no money. I have lots of money but no lemons. Because you have so many lemons, you don't value any given lemon very highly. Because I have so many dollars, I don't value any given dollar very highly.

So yeah, we don't value money equally... therefore what? We shouldn't trade? How absurd would that be? Of course we should trade. That's the entire point of trading. You give up something you value less (lemons) in order to get something you value more (money). I give up something that I value less (money) in order to get something that I value more (lemons). It's a mutually beneficial trade. We're both better off as a result. Now we both have lemons and money.

[The Two Jerseys]

Anytime...

Just in case you missed it, when Michelle auctions the quotes, the kids who win them get to decide which class/school goods their money gets spent on.

So not only are you not solving the free-rider "problem", you're making it even easier to be a free-rider...