The Yard Sale Model
December 21, 2022 12:45 PM Subscribe
Let's take a second to appreciate what just happened.
"This is the crux of the Yard-sale model. In a free market, one person ends up with all of the wealth - completely by chance."
"We believe that this purely analytical approach, which resembles an x-ray in that it is used not so much to represent the messiness of the real world as to strip it away and reveal the underlying skeleton, provides deep insight into the forces acting to increase poverty and inequality today. - Bruce M. Boghosian"
- You lost the first game.
- You won the second game.
- So you've won 50% of your games.
- But you have less money that you started with.
"This is the crux of the Yard-sale model. In a free market, one person ends up with all of the wealth - completely by chance."
"We believe that this purely analytical approach, which resembles an x-ray in that it is used not so much to represent the messiness of the real world as to strip it away and reveal the underlying skeleton, provides deep insight into the forces acting to increase poverty and inequality today. - Bruce M. Boghosian"
I suppose this might tell us something interesting if the typical economic interaction between people were a coinflip (or some other random procedure), and that the winner then received some kind of zero-sum payment from the loser. But it doesn't tell us anything interesting, because that is not a typical economic interaction. I appreciate that, at the end of such a series of interactions, you'll have unequal distribution. So what? The model has nothing to do with the reality we live in.
Consider a collection of more typical economic interactions between people. Person A goes to a workplace and produces something; at the end of the week, Person A gets paid. Person B goes to the grocery store and shops; Person B walks out with $50 of groceries and $50 less cash. Person C goes to the barber; Person C walks out with a haircut, but is now $25 poorer.
What is amazing about this is that all of the actors are now better off! Person A, B, and C are all wealthier, as are the owners of the workplace, the grocery store, and the barber shop. Every one of them is better off than before. It is amazing when you think about it.
posted by PaulVario at 1:11 PM on December 21, 2022 [5 favorites]
Consider a collection of more typical economic interactions between people. Person A goes to a workplace and produces something; at the end of the week, Person A gets paid. Person B goes to the grocery store and shops; Person B walks out with $50 of groceries and $50 less cash. Person C goes to the barber; Person C walks out with a haircut, but is now $25 poorer.
What is amazing about this is that all of the actors are now better off! Person A, B, and C are all wealthier, as are the owners of the workplace, the grocery store, and the barber shop. Every one of them is better off than before. It is amazing when you think about it.
posted by PaulVario at 1:11 PM on December 21, 2022 [5 favorites]
This is interesting! (And reminiscent of Stephen Jay Gould's Full House, which speculates about the possible consequences of this kind of self-organized asymmetry for everything from baseball to evolution.)
However, the following paragraphs make me wonder how well the author understood what they were writing about:
If you play enough rounds, both players will win about half the games. But the poorer player will lose most of their money.
Meanwhile, the richer player will gain money. That's because, from their perspective, every game they lose means they have an opportunity to win it back – and then some – in the next coin flip. Every game they win means, no matter what happens in the next coin flip, they'll still be at a net-plus.
This is absolutely wrong. Whether you lose a coin flip then win a coin flip, or win then lose, the results are the same: you come out behind. Everyone in this game is subject to the same skewed distribution of long-term outcomes, with a high chance of losing money and a low chance of gaining money (potentially gaining much more than the losers are able to lose, hence the skew). In general, the players who happen to be rich at any stage do not have an advantage in this model -- most of them will also be fractally ruined as wealth accumulates in fewer and fewer hands. The model demonstrates how a simple mechanism can create inequality "out of nowhere", but it fails to capture anything like the aspects of capitalism whereby the rich can leverage their wealth to get richer.
posted by aws17576 at 1:14 PM on December 21, 2022 [6 favorites]
However, the following paragraphs make me wonder how well the author understood what they were writing about:
If you play enough rounds, both players will win about half the games. But the poorer player will lose most of their money.
Meanwhile, the richer player will gain money. That's because, from their perspective, every game they lose means they have an opportunity to win it back – and then some – in the next coin flip. Every game they win means, no matter what happens in the next coin flip, they'll still be at a net-plus.
This is absolutely wrong. Whether you lose a coin flip then win a coin flip, or win then lose, the results are the same: you come out behind. Everyone in this game is subject to the same skewed distribution of long-term outcomes, with a high chance of losing money and a low chance of gaining money (potentially gaining much more than the losers are able to lose, hence the skew). In general, the players who happen to be rich at any stage do not have an advantage in this model -- most of them will also be fractally ruined as wealth accumulates in fewer and fewer hands. The model demonstrates how a simple mechanism can create inequality "out of nowhere", but it fails to capture anything like the aspects of capitalism whereby the rich can leverage their wealth to get richer.
posted by aws17576 at 1:14 PM on December 21, 2022 [6 favorites]
The real sin of capitalism is that the ten wealthiest people in the world aren't shot into the sun each year as a warning to everyone else to stop hoarding capital.
posted by seanmpuckett at 1:18 PM on December 21, 2022 [27 favorites]
posted by seanmpuckett at 1:18 PM on December 21, 2022 [27 favorites]
Actually, I spoke too soon (and too stridently). The two-player dynamics are as the author described -- they just aren't necessarily a good representation of the many-player version of the model. Mea culpa.
posted by aws17576 at 1:23 PM on December 21, 2022 [7 favorites]
posted by aws17576 at 1:23 PM on December 21, 2022 [7 favorites]
What is amazing about this is that all of the actors are now better off!
I think the demonstration isn't about capitalism -- where two parties agree to a transaction that benefits them equally -- and more about "wealth" as having money beyond the cost of living and capital as "something I own which makes me more money". When you are wealthy enough to have money in the stock market, it's money locked up someplace with the hopes it makes you profit, and you assume a certain amount of risk that it will become worth less in the future. The coinflips represent a known event causing the up or down, so it's more measurable in time, versus the "a butterfly flapped its wings, where did my retirement go" event. This is showing that even in a truly random situation where money goes up or down but you never lose it all, only a certain percent of the total owned, funnels more money to people who can afford to take higher risks, completely removed from any "skill" or "knowledge" that the rich person believes they might have resulting in their wealth.
Maybe I missed why it's called a "yard sale" model because yard sales aren't about profiting, it's about getting rid of stuff with a small amount of return, more like the bundling of junk mortgages for pennies on the dollar, which people bought hoping that there's something amazingly profitable hidden in....oh, wait, maybe I do get it now.
posted by AzraelBrown at 1:31 PM on December 21, 2022 [7 favorites]
I think the demonstration isn't about capitalism -- where two parties agree to a transaction that benefits them equally -- and more about "wealth" as having money beyond the cost of living and capital as "something I own which makes me more money". When you are wealthy enough to have money in the stock market, it's money locked up someplace with the hopes it makes you profit, and you assume a certain amount of risk that it will become worth less in the future. The coinflips represent a known event causing the up or down, so it's more measurable in time, versus the "a butterfly flapped its wings, where did my retirement go" event. This is showing that even in a truly random situation where money goes up or down but you never lose it all, only a certain percent of the total owned, funnels more money to people who can afford to take higher risks, completely removed from any "skill" or "knowledge" that the rich person believes they might have resulting in their wealth.
Maybe I missed why it's called a "yard sale" model because yard sales aren't about profiting, it's about getting rid of stuff with a small amount of return, more like the bundling of junk mortgages for pennies on the dollar, which people bought hoping that there's something amazingly profitable hidden in....oh, wait, maybe I do get it now.
posted by AzraelBrown at 1:31 PM on December 21, 2022 [7 favorites]
@PaulVario - I'm not sure that your example set of interactions can really be said to be any less 'spherical cow' than the one in the article. Sure you can tip a toy model to point whatever direction you want by selecting which parts get modeled & which externalities get excluded, but that doesn't make it any truer to the world. You've taken it as assumed that all actors in question are better off & wealthier, and written your example to match.
The model in the article is pretty up-front about it not being intended to model 'typical economic interactions'. It's modeling friction/transaction costs, & how iteration can produce complex behavior.
To borrow from an older "play with a model to see what can emerge", Nicky Case's Parable of the Polygons. It doesn't attempt to claim that people choose where they live according to as simple rules as are presented, but it *does* illustrate how Thomas Schelling's 'tipping point' model of segregation can emerge from simple individual biases.
Back to the original model, it demonstrates a way that simple rules can emerge into complex behavior, and suggests something which might mitigate that if similar behavior exists in the world. Or, the idea that exponential inequality can emerge without requiring someone 'planning' it. And that that's not how it has to be.
Which, y'know, comes up a *lot* in Adam Smith's writings. "But the rate of profit does not, like rent and wages, rise with the prosperity and fall with the declension of the society. On the contrary, it is naturally low in rich and high in poor countries, and it is always highest in the countries which are going fastest to ruin. The interest of this third order, therefore, has not the same connection with the general interest of the society as that of the other two."
I'm not saying Adam Smith is the be-all of theory about capitalism, but I am saying it goes back pretty far as an observation, and avoiding the implications of one model in favor of another isn't actually making an argument. It's just going "I like my numbers better". If you feel that your model is a 'truer' one, I'd love to see you show your work.
posted by CrystalDave at 1:33 PM on December 21, 2022 [10 favorites]
The model in the article is pretty up-front about it not being intended to model 'typical economic interactions'. It's modeling friction/transaction costs, & how iteration can produce complex behavior.
To borrow from an older "play with a model to see what can emerge", Nicky Case's Parable of the Polygons. It doesn't attempt to claim that people choose where they live according to as simple rules as are presented, but it *does* illustrate how Thomas Schelling's 'tipping point' model of segregation can emerge from simple individual biases.
Back to the original model, it demonstrates a way that simple rules can emerge into complex behavior, and suggests something which might mitigate that if similar behavior exists in the world. Or, the idea that exponential inequality can emerge without requiring someone 'planning' it. And that that's not how it has to be.
Which, y'know, comes up a *lot* in Adam Smith's writings. "But the rate of profit does not, like rent and wages, rise with the prosperity and fall with the declension of the society. On the contrary, it is naturally low in rich and high in poor countries, and it is always highest in the countries which are going fastest to ruin. The interest of this third order, therefore, has not the same connection with the general interest of the society as that of the other two."
I'm not saying Adam Smith is the be-all of theory about capitalism, but I am saying it goes back pretty far as an observation, and avoiding the implications of one model in favor of another isn't actually making an argument. It's just going "I like my numbers better". If you feel that your model is a 'truer' one, I'd love to see you show your work.
posted by CrystalDave at 1:33 PM on December 21, 2022 [10 favorites]
I found this fascinating, and really helpful to both how I think about wealth inequality, and how I might talk about it with others.
(And: I LOVE The Pudding, and am so glad they keep putting out interesting and well-designed pieces like this.)
The one thing I wish they might have changed: putting the .5% redistribution into real numbers.
In the redistribution simulation, the richest topped out at about $6700 (hm, the second time I ran it, it was $7000), and the poorest was at $41 (pretty sure it was $64 the first time I ran it).
So, some representative numbers - the redistribution tax would be:
wealth tax
$50.00 $0.25
$100.00 $0.50
$500.00 $2.50
$1,000.00 $5.00
$2,000.00 $10.00
$4,000.00 $20.00
$6,000.00 $30.00
$7,000.00 $35.00
$50,000.00 $250.00
$75,000.00 $375.00
$100,000.00 $500.00
I actually know some folks with INCOMES - not wealth - in the $50,000-$100,000 range, and those redistribution amounts are less than what they give voluntarily in actual charity (food banks, other direct needs support) most years.
I also love that the piece has links to the original papers at the end. (Note: the Anirban Chakraborti paper has the right link within the article but the wrong link at the end - it should go to Distributions of Money in Model Markets of Economy.)
This is a really interesting read, and I'm glad to have this to think about. Thank you so much for posting it, mhoye!
posted by kristi at 1:36 PM on December 21, 2022 [2 favorites]
(And: I LOVE The Pudding, and am so glad they keep putting out interesting and well-designed pieces like this.)
The one thing I wish they might have changed: putting the .5% redistribution into real numbers.
In the redistribution simulation, the richest topped out at about $6700 (hm, the second time I ran it, it was $7000), and the poorest was at $41 (pretty sure it was $64 the first time I ran it).
So, some representative numbers - the redistribution tax would be:
wealth tax
$50.00 $0.25
$100.00 $0.50
$500.00 $2.50
$1,000.00 $5.00
$2,000.00 $10.00
$4,000.00 $20.00
$6,000.00 $30.00
$7,000.00 $35.00
$50,000.00 $250.00
$75,000.00 $375.00
$100,000.00 $500.00
I actually know some folks with INCOMES - not wealth - in the $50,000-$100,000 range, and those redistribution amounts are less than what they give voluntarily in actual charity (food banks, other direct needs support) most years.
I also love that the piece has links to the original papers at the end. (Note: the Anirban Chakraborti paper has the right link within the article but the wrong link at the end - it should go to Distributions of Money in Model Markets of Economy.)
This is a really interesting read, and I'm glad to have this to think about. Thank you so much for posting it, mhoye!
posted by kristi at 1:36 PM on December 21, 2022 [2 favorites]
In this model the rich get richer because they are by definition playing against poorer people, and the wager will be the lesser of 20% of each player's wealth. Rich people will lose less than they initially won, allowing them to stay net positive over time as their opponents get poorer. They only start to lose when they play people richer than they are.
If people bet the same amount every time regardless of their net worth this sort of inequality wouldn't manifest. A lot of people would go broke, but there would be no super-rich people.
posted by grumpybear69 at 1:37 PM on December 21, 2022 [1 favorite]
If people bet the same amount every time regardless of their net worth this sort of inequality wouldn't manifest. A lot of people would go broke, but there would be no super-rich people.
posted by grumpybear69 at 1:37 PM on December 21, 2022 [1 favorite]
... and another thought:
For me, the value of the piece is simply illustrating that "intuitive common sense" is wrong: I think a lot of us share the thinking that, even if we lose out sometimes, it'll all even out; we'll win some, too ... and then when that DOESN'T happen, when we end up struggling and broke, it's easy to feel both shame and betrayal - neither of which build healthy lives, healthy communities, or healthy ways of understanding the economic structures we support for our nation.
Getting to see that even in completely random transactions, some people end up exceedingly wealthy while the majority end up massively worse off - that is a valuable illustration of mistaken thinking, and an invitation to explore how that works and how we might mitigate it.
posted by kristi at 1:40 PM on December 21, 2022 [11 favorites]
For me, the value of the piece is simply illustrating that "intuitive common sense" is wrong: I think a lot of us share the thinking that, even if we lose out sometimes, it'll all even out; we'll win some, too ... and then when that DOESN'T happen, when we end up struggling and broke, it's easy to feel both shame and betrayal - neither of which build healthy lives, healthy communities, or healthy ways of understanding the economic structures we support for our nation.
Getting to see that even in completely random transactions, some people end up exceedingly wealthy while the majority end up massively worse off - that is a valuable illustration of mistaken thinking, and an invitation to explore how that works and how we might mitigate it.
posted by kristi at 1:40 PM on December 21, 2022 [11 favorites]
This is compounding at work right?
Well, no, its basically a geometric mean. Retirement investment literature calls it "sequence of return" risk, because losing 10 percent then gaining ten percent is different than the reverse (gaining 10 percent then losing 10 percent). This is why diversification and passive index investing is useful, it reduces the variance of personal outcomes.
However, I don't think this Yard Sale model has much to do with how oligarchs form. Most absurdly rich people I can think of seem to be all in one investment going really, really well, for at least a decade. Its like Larry Page bet 20 percent and won 2 million percent back. Even the VCs who do engage in repeated investing transactions are geared towards finding one really good investment rather than a string of lucky wins.
posted by pwnguin at 1:53 PM on December 21, 2022
Well, no, its basically a geometric mean. Retirement investment literature calls it "sequence of return" risk, because losing 10 percent then gaining ten percent is different than the reverse (gaining 10 percent then losing 10 percent). This is why diversification and passive index investing is useful, it reduces the variance of personal outcomes.
However, I don't think this Yard Sale model has much to do with how oligarchs form. Most absurdly rich people I can think of seem to be all in one investment going really, really well, for at least a decade. Its like Larry Page bet 20 percent and won 2 million percent back. Even the VCs who do engage in repeated investing transactions are geared towards finding one really good investment rather than a string of lucky wins.
posted by pwnguin at 1:53 PM on December 21, 2022
This supposes a homogeneous population where everyone has a chance to win. In reality race, gender, sexuality, faith, education - all the other factors of inequality - mean that absent the luck of Rosencrantz, some people will never win.
posted by adept256 at 1:54 PM on December 21, 2022
posted by adept256 at 1:54 PM on December 21, 2022
This is compounding at work right?
It's more about how regressive it is to use the same percentage for everyone.
Losing 20% of your money means something entirely different to the millionaire and to the person with $10,000.
posted by straight at 1:58 PM on December 21, 2022 [2 favorites]
It's more about how regressive it is to use the same percentage for everyone.
Losing 20% of your money means something entirely different to the millionaire and to the person with $10,000.
posted by straight at 1:58 PM on December 21, 2022 [2 favorites]
Okay, just one more comment:
I clicked into the Scientific American article, "Is Inequality Inevitable? Wealth naturally trickles up in free-market economies, model suggests", By Bruce M. Boghosian, which contains this line:
So, in some aspects, this approach has a lot to do with the real world.
posted by kristi at 1:58 PM on December 21, 2022 [5 favorites]
I clicked into the Scientific American article, "Is Inequality Inevitable? Wealth naturally trickles up in free-market economies, model suggests", By Bruce M. Boghosian, which contains this line:
And astonishingly, [adding the redistribution of wealth to the original model] enabled our model to match empirical data on U.S. and European wealth distribution between 1989 and 2016 to better than 2 percent.It shows charts overlaying their model on actual data for the US in 1989 and 2016, and Germany and Greece in 2010: they're nearly identical.
So, in some aspects, this approach has a lot to do with the real world.
posted by kristi at 1:58 PM on December 21, 2022 [5 favorites]
Getting to see that even in completely random transactions, some people end up exceedingly wealthy while the majority end up massively worse off - that is a valuable illustration of mistaken thinking, and an invitation to explore how that works and how we might mitigate it.
like why socialism is a mitigating response to capitalism's winner-take-all endowment effects, if not (an overcorrective) leveling the playing field altogether -- and a brake on fascism.
posted by kliuless at 2:00 PM on December 21, 2022 [3 favorites]
like why socialism is a mitigating response to capitalism's winner-take-all endowment effects, if not (an overcorrective) leveling the playing field altogether -- and a brake on fascism.
posted by kliuless at 2:00 PM on December 21, 2022 [3 favorites]
Ha, I took a Lie Groups class with Bruce Boghosian when I was like 19 and woefully under prepared for the material.
posted by kaibutsu at 2:34 PM on December 21, 2022 [1 favorite]
posted by kaibutsu at 2:34 PM on December 21, 2022 [1 favorite]
I played the two player game as the poorer person and won pretty handily. Not to overplay this but they do say "if you play enough rounds, both players will win about half the games. But the poorer player will lose most of their money." which is not equivocal.
Liked the article otherwise.
posted by Rumple at 2:37 PM on December 21, 2022 [1 favorite]
Liked the article otherwise.
posted by Rumple at 2:37 PM on December 21, 2022 [1 favorite]
The problem for me is that the yard-sale model explains (1) what doesn't need to be explained and also explains (2) things that aren't actually true. It also (3) uses a very weird model of economic life that seems to have no relation to reality.
I mean, does it really need to be explained that if everybody starts with a bank account of $X, eventually some people will have more money than others because of various spending and consumption patterns? You don't need to pretend that everybody's economic interactions are driven by a bunch of games of Liar's Poker with peculiar and subjective betting limits in order to derive this result!
I think quotes like "We believe that this purely analytical approach, which resembles an x-ray in that it is used not so much to represent the messiness of the real world as to strip it away and reveal the underlying skeleton, provides deep insight into the forces acting to increase poverty and inequality today" reveal something important: in the real world, there is no skeleton of Liar's Poker parties that represent economic life in any meaningful way!
That's why it's silly for the author to write both that 20% of families own 70% of wealth and that the Yard-Sale model predicts you're going to end up with just one super-rich person. The model doesn't explain the outcome.
I appreciate that lots of people in the comments are struggling to explain the "underlying skeleton." Is it just investments? Is it compound interest? What this demonstrates is that the article that's supposed to explain the significance of the yard-sale dynamic isn't at all clear. I think that has something to do with the fact that the yard-sale doesn't really explain much of anything.
posted by PaulVario at 2:56 PM on December 21, 2022
I mean, does it really need to be explained that if everybody starts with a bank account of $X, eventually some people will have more money than others because of various spending and consumption patterns? You don't need to pretend that everybody's economic interactions are driven by a bunch of games of Liar's Poker with peculiar and subjective betting limits in order to derive this result!
I think quotes like "We believe that this purely analytical approach, which resembles an x-ray in that it is used not so much to represent the messiness of the real world as to strip it away and reveal the underlying skeleton, provides deep insight into the forces acting to increase poverty and inequality today" reveal something important: in the real world, there is no skeleton of Liar's Poker parties that represent economic life in any meaningful way!
That's why it's silly for the author to write both that 20% of families own 70% of wealth and that the Yard-Sale model predicts you're going to end up with just one super-rich person. The model doesn't explain the outcome.
I appreciate that lots of people in the comments are struggling to explain the "underlying skeleton." Is it just investments? Is it compound interest? What this demonstrates is that the article that's supposed to explain the significance of the yard-sale dynamic isn't at all clear. I think that has something to do with the fact that the yard-sale doesn't really explain much of anything.
posted by PaulVario at 2:56 PM on December 21, 2022
> PaulVario: "But it doesn't tell us anything interesting, because that is not a typical economic interaction."
Huh, I didn't interpret this model as mainly being about the commerce side (i.e.: the exchange of money for goods and services) of the economy. I thought this was about the finance side of the economy (i.e.: let's me and you bet on whether this stock/bond/other financial instrument is gonna go up or down).
posted by mhum at 3:07 PM on December 21, 2022 [5 favorites]
Huh, I didn't interpret this model as mainly being about the commerce side (i.e.: the exchange of money for goods and services) of the economy. I thought this was about the finance side of the economy (i.e.: let's me and you bet on whether this stock/bond/other financial instrument is gonna go up or down).
posted by mhum at 3:07 PM on December 21, 2022 [5 favorites]
I played the two player game as the poorer person and won pretty handily.
Indeed, this can happen, and it's even possible to work out the likelihood of its happening by a simple principle!
The yard sale model is a martingale, a process in which the "expected value" of each player's holdings at some time in the future is equal to their present holdings. I put expected value in scare quotes because it's a term of art and arguably a misnomer; it's the probabilistically weighted average of all possible outcomes, but is not necessarily in itself an outcome that should be expected. What can actually be "expected" in this game, in the long run, is that one player will go broke.
Which player? Well, the expected value of your holdings after, say, a thousand coin flips must be $100, equal to the amount you had at the outset. Given that you will actually almost certainly have (nearly) $0 or $1100, the probability attached to the latter outcome, to make the weighted average come out as we know it must, must be 1/11.
The fact that the yard sale model is a martingale is part of what makes it compelling -- it's a fair game in a certain sense, just one in which the winner tends to take all. (Which perhaps makes it an unwise game to play regardless of whether it is "fair". I've never felt that the problem with the lottery is that the house takes a cut; if the lottery paid out double what it took in, but all to one person, I'd still be reluctant to play, though my opinion would change if I could afford to buy all the tickets.)
posted by aws17576 at 4:46 PM on December 21, 2022 [5 favorites]
Indeed, this can happen, and it's even possible to work out the likelihood of its happening by a simple principle!
The yard sale model is a martingale, a process in which the "expected value" of each player's holdings at some time in the future is equal to their present holdings. I put expected value in scare quotes because it's a term of art and arguably a misnomer; it's the probabilistically weighted average of all possible outcomes, but is not necessarily in itself an outcome that should be expected. What can actually be "expected" in this game, in the long run, is that one player will go broke.
Which player? Well, the expected value of your holdings after, say, a thousand coin flips must be $100, equal to the amount you had at the outset. Given that you will actually almost certainly have (nearly) $0 or $1100, the probability attached to the latter outcome, to make the weighted average come out as we know it must, must be 1/11.
The fact that the yard sale model is a martingale is part of what makes it compelling -- it's a fair game in a certain sense, just one in which the winner tends to take all. (Which perhaps makes it an unwise game to play regardless of whether it is "fair". I've never felt that the problem with the lottery is that the house takes a cut; if the lottery paid out double what it took in, but all to one person, I'd still be reluctant to play, though my opinion would change if I could afford to buy all the tickets.)
posted by aws17576 at 4:46 PM on December 21, 2022 [5 favorites]
It really is some truly obtuse behaviour to turn up here and not understand what’s showed in the demonstration and to attempt to refute it with a weird misexplanation of comparative advantage.
But if there’s one thing I’ve learned, it’s that every time you give a well worked example of a market failure, there will be someone who comes along and patronisingly explains to you what a market is.
posted by ambrosen at 4:51 PM on December 21, 2022 [13 favorites]
But if there’s one thing I’ve learned, it’s that every time you give a well worked example of a market failure, there will be someone who comes along and patronisingly explains to you what a market is.
posted by ambrosen at 4:51 PM on December 21, 2022 [13 favorites]
Thanks ambrosen: I’m not sure if you understand what “comparative advantage” is, but it hasn’t been discussed here: you’re apparently confusing it with what’s conventionally called “the double coincidence of wants.”
I wouldn’t call what’s under discussion “a well-worked example of a market failure,” either, largely because there is no real-world analogue to a universe of economic transactions that consists solely of people making up-down bets with zero expected value. In order for models to say anything interesting, they first have to have some relationship to the real world.
But, go ahead, feel free to divest yourself of the sneers and insults: explain to us “what’s showed in the demonstration.” I’d suggest “zero” or “not much,” but presumably your estimate is higher.
posted by PaulVario at 5:39 PM on December 21, 2022
I wouldn’t call what’s under discussion “a well-worked example of a market failure,” either, largely because there is no real-world analogue to a universe of economic transactions that consists solely of people making up-down bets with zero expected value. In order for models to say anything interesting, they first have to have some relationship to the real world.
But, go ahead, feel free to divest yourself of the sneers and insults: explain to us “what’s showed in the demonstration.” I’d suggest “zero” or “not much,” but presumably your estimate is higher.
posted by PaulVario at 5:39 PM on December 21, 2022
(Note- I started typing this before seeing your most recent comment, which I now see on "preview", but I think this still applies)
@PaulVario - I think you are overestimating the average level of understanding about how markets work.
Many, many people believe that capitalist societies are inherently meritocracies. This line of thinking leads to the notion that rich people are rich because they are smart, and poor people are poor because they are foolish.
It leads to phrases like "We all have the same 24 hours in a day" or "So are you saying the government should just take everyone's money?"
I'm generalizing here, but these lines of thinking are often paired with a zero-sum approach to things like taxes. See also: "Why should my money be given to people who made bad decisions?"
The essay and the browser simulations do a good job of demonstrating how, even in an artificially balanced system, a series of zero-sum exchanges spread over time will not cause things to "even out" for everybody. In fact, even in a series of seemingly equal zero-sum exchanges, random chance can create supreme wealth inequality.
It also does a very strong job of showing have extremely small and reasonable taxation can be life-changing for those at the bottom, while still leaving those at the top with extremely comfortable margins.
Your barbershop example is a fantastic demonstration of how exchanges can be mutually beneficial even if they cost money. To put it another way: in the right contexts, capitalism is not zero-sum. That alone, I think, puts you outside of the target demographic for the link.
Does that make sense?
posted by ®@ at 5:53 PM on December 21, 2022 [4 favorites]
@PaulVario - I think you are overestimating the average level of understanding about how markets work.
Many, many people believe that capitalist societies are inherently meritocracies. This line of thinking leads to the notion that rich people are rich because they are smart, and poor people are poor because they are foolish.
It leads to phrases like "We all have the same 24 hours in a day" or "So are you saying the government should just take everyone's money?"
I'm generalizing here, but these lines of thinking are often paired with a zero-sum approach to things like taxes. See also: "Why should my money be given to people who made bad decisions?"
The essay and the browser simulations do a good job of demonstrating how, even in an artificially balanced system, a series of zero-sum exchanges spread over time will not cause things to "even out" for everybody. In fact, even in a series of seemingly equal zero-sum exchanges, random chance can create supreme wealth inequality.
It also does a very strong job of showing have extremely small and reasonable taxation can be life-changing for those at the bottom, while still leaving those at the top with extremely comfortable margins.
Your barbershop example is a fantastic demonstration of how exchanges can be mutually beneficial even if they cost money. To put it another way: in the right contexts, capitalism is not zero-sum. That alone, I think, puts you outside of the target demographic for the link.
Does that make sense?
posted by ®@ at 5:53 PM on December 21, 2022 [4 favorites]
People on top will always turn the invisible hand into a ratchet, I guess.
I don't have much time for people who are sealioning about why that isn't so.
posted by ambrosen at 5:55 PM on December 21, 2022 [2 favorites]
I don't have much time for people who are sealioning about why that isn't so.
posted by ambrosen at 5:55 PM on December 21, 2022 [2 favorites]
there is no real-world analogue to a universe of economic transactions that consists solely of people making up-down bets with zero expected value
Sure there is: stock markets.
It's not the same thing as what you're describing, but it is analogous in many ways. The thing with modelling is that you're inherently making simplifications, reducing things down to a number, a mathematical interaction or function. It inherently won't be the same thing. And frankly, this example is much more closely representative of the thing its modelling - not that that is necessarily the goal - than so much of what goes on in a lot of economic modelling.
posted by Dysk at 6:05 PM on December 21, 2022 [4 favorites]
Sure there is: stock markets.
It's not the same thing as what you're describing, but it is analogous in many ways. The thing with modelling is that you're inherently making simplifications, reducing things down to a number, a mathematical interaction or function. It inherently won't be the same thing. And frankly, this example is much more closely representative of the thing its modelling - not that that is necessarily the goal - than so much of what goes on in a lot of economic modelling.
posted by Dysk at 6:05 PM on December 21, 2022 [4 favorites]
To put it another way: in my opinion, what the demonstration shows is that "common sense" conclusions can be very wrong when applied to systems like markets.
This is especially important information because so many of the arguments we hear about economic policy are framed as "well, it's just common sense!"
The benefits of progressive taxation are very measurable but they *feel* wrong.
How does the government stealing money from rich people and giving it to poor people make things any better for any of us?
That's a really good question! The answer is, frustratingly, "well, it's very complicated, but if we only steal a little bit from people who can afford it things actually appear to work out very well for a lot of people and the rich people still have enough money to take risks and buy fun rich person things. But there are a lot of numbers and stuff so you just have to trust us."
That is a really bad and unsatisfying answer!
These kinds of thought experiments help to break through that barrier. On preview: Also what Dysk said!
Again, it sounds like this is stuff you already know, which is why I say you are probably not the target audience for these links.
As to why you are receiving flack in the thread: I don't think you've done anything wrong, but I would say that your comments sound a little bit like a mathematics graduate student throwing shade at a blog post for explaining first-grade mathematics. Sure, that stuff may be obvious to you, but it's not obvious to everybody.
posted by ®@ at 6:06 PM on December 21, 2022 [3 favorites]
This is especially important information because so many of the arguments we hear about economic policy are framed as "well, it's just common sense!"
The benefits of progressive taxation are very measurable but they *feel* wrong.
How does the government stealing money from rich people and giving it to poor people make things any better for any of us?
That's a really good question! The answer is, frustratingly, "well, it's very complicated, but if we only steal a little bit from people who can afford it things actually appear to work out very well for a lot of people and the rich people still have enough money to take risks and buy fun rich person things. But there are a lot of numbers and stuff so you just have to trust us."
That is a really bad and unsatisfying answer!
These kinds of thought experiments help to break through that barrier. On preview: Also what Dysk said!
Again, it sounds like this is stuff you already know, which is why I say you are probably not the target audience for these links.
As to why you are receiving flack in the thread: I don't think you've done anything wrong, but I would say that your comments sound a little bit like a mathematics graduate student throwing shade at a blog post for explaining first-grade mathematics. Sure, that stuff may be obvious to you, but it's not obvious to everybody.
posted by ®@ at 6:06 PM on December 21, 2022 [3 favorites]
I think the demonstration isn't about capitalism -- where two parties agree to a transaction that benefits them equally -- and more about "wealth" as having money beyond the cost of living and capital as "something I own which makes me more money".
But that's what capitalism is...capitalists have capital. The first thing, mutually beneficial transactions, is just a market which isn't inherently capitalist. Markets pre-date capitalism by several centuries at least. The latter thing, which you call wealth, is textbook capitalism distilled into a brief sentence.
posted by asnider at 7:09 PM on December 21, 2022 [6 favorites]
But that's what capitalism is...capitalists have capital. The first thing, mutually beneficial transactions, is just a market which isn't inherently capitalist. Markets pre-date capitalism by several centuries at least. The latter thing, which you call wealth, is textbook capitalism distilled into a brief sentence.
posted by asnider at 7:09 PM on December 21, 2022 [6 favorites]
Follow the Money is the research paper referenced in the FPP, authored by Brian Hayes, the one who first called it the Yard-sale model. Has some interesting variations on the model and a discussion on the implications of the model vis-a-vis markets etc.
posted by storybored at 8:55 PM on December 21, 2022 [2 favorites]
posted by storybored at 8:55 PM on December 21, 2022 [2 favorites]
Really liked this post, thank you for sharing it
posted by lazaruslong at 2:58 AM on December 22, 2022
posted by lazaruslong at 2:58 AM on December 22, 2022
To put it another way: in the right contexts, capitalism is not zero-sum.
You're not talking about capitalism, you're talking about free exchange.
Capitalism exists on top of free exchange as a method of deciding who gets the profit the factory generates. You can still have free exchange even if by law and policy the factory's profit goes to the people who work there. Or, you can go a long way out towards the kind of dystopia Hayek envisioned and still have free exchange.
I am rusty on this stuff but it goes way back to the old theorems of welfare economics. We can pick pretty much any point we like on the production possibility frontier. And we don't need to worry very much about the initial allocation of resources it requires to get there, since free exchange will at least move us to one or another pareto optimal distribution. All we need to do is figure out where on the PPF we want to move to, and then Arrow ruined it all.
What is amazing about this is that all of the actors are now better off! Person A, B, and C are all wealthier, as are the owners of the workplace, the grocery store, and the barber shop.
Better off != wealthier. Person B is down $50 and has now-second-hand groceries the market values at less than $50, but which they value higher than that. Consumer surplus is great but you can't invest it.
posted by GCU Sweet and Full of Grace at 4:44 AM on December 22, 2022 [4 favorites]
You're not talking about capitalism, you're talking about free exchange.
Capitalism exists on top of free exchange as a method of deciding who gets the profit the factory generates. You can still have free exchange even if by law and policy the factory's profit goes to the people who work there. Or, you can go a long way out towards the kind of dystopia Hayek envisioned and still have free exchange.
I am rusty on this stuff but it goes way back to the old theorems of welfare economics. We can pick pretty much any point we like on the production possibility frontier. And we don't need to worry very much about the initial allocation of resources it requires to get there, since free exchange will at least move us to one or another pareto optimal distribution. All we need to do is figure out where on the PPF we want to move to, and then Arrow ruined it all.
What is amazing about this is that all of the actors are now better off! Person A, B, and C are all wealthier, as are the owners of the workplace, the grocery store, and the barber shop.
Better off != wealthier. Person B is down $50 and has now-second-hand groceries the market values at less than $50, but which they value higher than that. Consumer surplus is great but you can't invest it.
posted by GCU Sweet and Full of Grace at 4:44 AM on December 22, 2022 [4 favorites]
The thing I really appreciate about the Yard Sale model is that when you sweep away the money--with all of its emotional and psychological entanglements--what you see is an engine for aggregating resources, and it's important to recognize those engines all around us, because they function not merely economically but politically. Every form of power has a tendency to centralize, and we end up with billionaires and aristocracies. Not because we inherently love billionaires (our weird emotional attachments to the rich come later in the process), and not because billionaires are great geniuses who know what they're doing (clearly), but because certain people find themselves on the right side of the engine. (It's like if you added one more step to the Yard Sale game: At the end of each round, the wealthier player gets to make a rule that the poorer player must follow.)
One thing PaulVario's model misses, I think, and GCU Sweet and Full of Grace's second-hand grocery comment illustrates this perfectly, is time. What happens next, after everyone goes home happy with their transactions? B, without $50, must ingest his groceries to generate the energy to sell his labor (since he can't sell his groceries), and suddenly we're in the first chapters of Capital.
posted by mittens at 5:56 AM on December 22, 2022 [1 favorite]
One thing PaulVario's model misses, I think, and GCU Sweet and Full of Grace's second-hand grocery comment illustrates this perfectly, is time. What happens next, after everyone goes home happy with their transactions? B, without $50, must ingest his groceries to generate the energy to sell his labor (since he can't sell his groceries), and suddenly we're in the first chapters of Capital.
posted by mittens at 5:56 AM on December 22, 2022 [1 favorite]
In this thread: a whole lot of people completely missing the point of TFA (which IMO is very fine indeed), namely that a system with such very simple, egalitarian-seeming rules invariably produces a wildly counterintuitive outcome over enough iterations.
To actually discuss TFA a bit: I'm curious about exactly how the redistribution is computed: is it X% of the person's income that turn, or X% of their winnings that turn?
Running the model repeatedly while bumping the "redistribution" by 0.5% (with resets after each bump), I found that a 2% setting on the "redistribution" parameter was enough to even out the outcomes to a range of < 40:1 (poorest: $181, richest: 3569).
Which jibes neatly with the intuition that the skyrocketing share of wealth of the super-wealthy has a whole hell of a lot to do with the changes in tax policy since 1980.
posted by Aardvark Cheeselog at 8:55 AM on December 22, 2022 [3 favorites]
To actually discuss TFA a bit: I'm curious about exactly how the redistribution is computed: is it X% of the person's income that turn, or X% of their winnings that turn?
Running the model repeatedly while bumping the "redistribution" by 0.5% (with resets after each bump), I found that a 2% setting on the "redistribution" parameter was enough to even out the outcomes to a range of < 40:1 (poorest: $181, richest: 3569).
Which jibes neatly with the intuition that the skyrocketing share of wealth of the super-wealthy has a whole hell of a lot to do with the changes in tax policy since 1980.
posted by Aardvark Cheeselog at 8:55 AM on December 22, 2022 [3 favorites]
And, reading a bit more carefully, I see it's "tax all players and redistribute," so it's wealth redistribution we're talking about.
Interestingly, after setting a redistribution % you need to run the model for several 1000 iterations before it stabilizes. At 0.5% redistribution, it seems to never really stabilize but to fluctuate between a 100:1 and 200:1 wealth distribution between the richest and poorest.
At 1% the ratio drops to around 50:1. With 2% redistribution the ratio converges to between 20 and 30 to 1.
To respond to somebody upthread who wanted to know what's the significance of the "wager amount," it serves to spread out the evolution of the model over time: it inversely correlates with how quickly the poorest have nothing, and directly correlates with how rich the winners get in the presence of redistribution. If you really, really insisted on forcing this model into something like a real-world interpretation, think of the wager as the amount invested and the rest as cash reserves, in a series of investments which each pay out (or not) in the course of a turn. In real life, then, since most people have basically no reserves, the wager should be close to 100%. But that makes the model collapse really quickly.
posted by Aardvark Cheeselog at 9:14 AM on December 22, 2022 [1 favorite]
Interestingly, after setting a redistribution % you need to run the model for several 1000 iterations before it stabilizes. At 0.5% redistribution, it seems to never really stabilize but to fluctuate between a 100:1 and 200:1 wealth distribution between the richest and poorest.
At 1% the ratio drops to around 50:1. With 2% redistribution the ratio converges to between 20 and 30 to 1.
To respond to somebody upthread who wanted to know what's the significance of the "wager amount," it serves to spread out the evolution of the model over time: it inversely correlates with how quickly the poorest have nothing, and directly correlates with how rich the winners get in the presence of redistribution. If you really, really insisted on forcing this model into something like a real-world interpretation, think of the wager as the amount invested and the rest as cash reserves, in a series of investments which each pay out (or not) in the course of a turn. In real life, then, since most people have basically no reserves, the wager should be close to 100%. But that makes the model collapse really quickly.
posted by Aardvark Cheeselog at 9:14 AM on December 22, 2022 [1 favorite]
Is this something else that would be solved with a 100% estate tax over the first $10MM?
posted by mikelieman at 1:36 PM on December 22, 2022
posted by mikelieman at 1:36 PM on December 22, 2022
It ought to be blindingly obvious to all that in any system where it's generally agreed that a "fair" return on any investment is a fixed percentage of the amount staked, regardless of absolute stake size, there exists a positive feedback guaranteeing that money will always gravitate toward those players who start out with more of it. Money makes money: the end.
It should also be blindingly obvious to all that once a person controls enough assets to generate at least enough income to live on just by continuing to exist, it makes no sense to ascribe virtue to that person on the basis of that wealth. But it isn't. Most people do not live that way; most people work for their daily bread and simply cannot wrap their heads around the idea that people who receive more bread than they do are not necessarily smarter than they are nor working harder than they do.
But people who live on the passive income generated by the assets they own are not "working hard for their money". They're just being given it by everybody else, no different from people who live on State welfare, except that the smarmy self-congratulatory fuck in a suit will almost always be getting way more than any welfare recipient has ever been given in the entire history of ever.
The ultra-rich are not our "betters". They're a pack of useless fucking bludgers and we'd all be better off for taxing the fuck out of them as the first port of call to fund anything that's in the public interest.
posted by flabdablet at 2:04 PM on December 22, 2022 [1 favorite]
It should also be blindingly obvious to all that once a person controls enough assets to generate at least enough income to live on just by continuing to exist, it makes no sense to ascribe virtue to that person on the basis of that wealth. But it isn't. Most people do not live that way; most people work for their daily bread and simply cannot wrap their heads around the idea that people who receive more bread than they do are not necessarily smarter than they are nor working harder than they do.
But people who live on the passive income generated by the assets they own are not "working hard for their money". They're just being given it by everybody else, no different from people who live on State welfare, except that the smarmy self-congratulatory fuck in a suit will almost always be getting way more than any welfare recipient has ever been given in the entire history of ever.
The ultra-rich are not our "betters". They're a pack of useless fucking bludgers and we'd all be better off for taxing the fuck out of them as the first port of call to fund anything that's in the public interest.
posted by flabdablet at 2:04 PM on December 22, 2022 [1 favorite]
I agree with your sentiments but I think you have perhaps overlooked why the dynamics of this model are interesting. Intuitively you might expect that if everyone is playing zero-expectation coinflips then basically everyone's wealth should random walk around, and the spread of wealth should kinda grow linearly over time like a bundle of random walks drifting apart.
But, then you say, well since everyone is risking a fraction of their wealth we should really be thinking about the random walk differently: suppose everyone risks 20% of their wealth, then when they win their wealth multiplies by 1.2, and when they lose their wealth multiplies by 0.8. So a person's wealth history is a sequence of multiplications like 1.2*1.2*0.8*1.2*0.8*0.8... etc. If we consider the log of their wealth then we see that ln(wealth) is a simple sum of terms like ln(1.2) + ln(1.2) + ln(0.8) + ln(1.2) + ... .
In other words, in log space a person's wealth is a random walk where each step they have an equal chance of getting +ln(1.2) or +ln(0.8). Now I've been hiding a curious fact: ln(1.2) ~= 0.1823, but ln(0.8) ~= -0.2231: these do not average to zero, in fact their average is negative. Considered in log space, the expected growth of your wealth for each coin flip is negative, and your log wealth steadily drifts downwards.
This is the kernel of what's interesting about this model: risking a fixed fraction of your wealth on this type of game will almost surely impoverish you in the long term, because the expected geometric growth rate is negative. But in this model players don't play against the house, they play against each other and the total wealth is conserved, so if every player 'almost surely' bankrupts themselves in the long term, where does the money go?
posted by Pyry at 3:11 PM on December 22, 2022 [4 favorites]
But, then you say, well since everyone is risking a fraction of their wealth we should really be thinking about the random walk differently: suppose everyone risks 20% of their wealth, then when they win their wealth multiplies by 1.2, and when they lose their wealth multiplies by 0.8. So a person's wealth history is a sequence of multiplications like 1.2*1.2*0.8*1.2*0.8*0.8... etc. If we consider the log of their wealth then we see that ln(wealth) is a simple sum of terms like ln(1.2) + ln(1.2) + ln(0.8) + ln(1.2) + ... .
In other words, in log space a person's wealth is a random walk where each step they have an equal chance of getting +ln(1.2) or +ln(0.8). Now I've been hiding a curious fact: ln(1.2) ~= 0.1823, but ln(0.8) ~= -0.2231: these do not average to zero, in fact their average is negative. Considered in log space, the expected growth of your wealth for each coin flip is negative, and your log wealth steadily drifts downwards.
This is the kernel of what's interesting about this model: risking a fixed fraction of your wealth on this type of game will almost surely impoverish you in the long term, because the expected geometric growth rate is negative. But in this model players don't play against the house, they play against each other and the total wealth is conserved, so if every player 'almost surely' bankrupts themselves in the long term, where does the money go?
posted by Pyry at 3:11 PM on December 22, 2022 [4 favorites]
I'd like some analysis of the upper half so my brain doesn't go "one person gets the most wealth by winning the most coin tosses." I figure it's the linear distance-from-zero vs logarithmic percentage staked that causes the imbalance in the random walk.
It's a percentage (proportional) to the distance already from zero, and nobody has said "don't deal with people richer than you" because their potential winnings eclipse your potential winnings. So the random win or loss is not merely a fair coin toss but is really "is your opponent richer than you and can you equal their stake?"
posted by k3ninho at 3:26 PM on December 22, 2022
It's a percentage (proportional) to the distance already from zero, and nobody has said "don't deal with people richer than you" because their potential winnings eclipse your potential winnings. So the random win or loss is not merely a fair coin toss but is really "is your opponent richer than you and can you equal their stake?"
posted by k3ninho at 3:26 PM on December 22, 2022
I think the resolution is, counterintuitively, that the benefit of wealth is getting to make small bets. If a player with 1,000,000 plays against one with 100, they'll wager min(0.2 * 1000000, 0.2 * 100) = 20. So the poor player is playing the fractional bet game with 20% of their wealth on the line (which we've established will almost certainly cause them to lose all their money sooner or later) while the rich player risks proportionally nothing.
So what happens is that one player gets a lead large enough that they effectively become 'the house' against which the other players gamble to their nearly guaranteed ruin.
posted by Pyry at 3:41 PM on December 22, 2022 [4 favorites]
So what happens is that one player gets a lead large enough that they effectively become 'the house' against which the other players gamble to their nearly guaranteed ruin.
posted by Pyry at 3:41 PM on December 22, 2022 [4 favorites]
I was really disappointed at first that this wasn't an actual yard sale RPG but TFA was interesting.
posted by urbanlenny at 3:49 PM on December 22, 2022
posted by urbanlenny at 3:49 PM on December 22, 2022
I suppose this might tell us something interesting if the typical economic interaction between people were a coinflip (or some other random procedure), and that the winner then received some kind of zero-sum payment from the loser. But it doesn't tell us anything interesting, because that is not a typical economic interaction. I appreciate that, at the end of such a series of interactions, you'll have unequal distribution. So what? The model has nothing to do with the reality we live in.
Consider all the complexities of water molecules and their interactions. They can store energy by wiggling their angle. They have charged parts that attract the charged parts of other water molecules. We know this. But despite that, the boring ideal gas model, which ignores all that, does a bang-on job of characterizing the behavior of water vapor. I don't believe we should say that a model is interesting only if it gets the microscopic particulars right. If it gets the macroscopic picture right, what can we learn from that? Are there microscopic particulars that turn out to not make as big a difference as we thought?
posted by a snickering nuthatch at 3:59 PM on December 22, 2022 [1 favorite]
Consider all the complexities of water molecules and their interactions. They can store energy by wiggling their angle. They have charged parts that attract the charged parts of other water molecules. We know this. But despite that, the boring ideal gas model, which ignores all that, does a bang-on job of characterizing the behavior of water vapor. I don't believe we should say that a model is interesting only if it gets the microscopic particulars right. If it gets the macroscopic picture right, what can we learn from that? Are there microscopic particulars that turn out to not make as big a difference as we thought?
posted by a snickering nuthatch at 3:59 PM on December 22, 2022 [1 favorite]
where does the money go?
That's what the article was REALLY bad at explaining.
Let's assume a 4 iteration game to make it simple, but this applies for the any n-iterations you like.
There are 16 different outcomes, if you start with $100, listed below.
$107 - 4 wins
$ 38 - 3 wins 1 loss
$ 38 - 3 wins 1 loss
$ 38 - 3 wins 1 loss
$ 38 - 3 wins 1 loss
$ (8) - 2 wins 2 losses
$ (8) - 2 wins 2 losses
$ (8) - 2 wins 2 losses
$ (8) - 2 wins 2 losses
$ (8) - 2 wins 2 losses
$ (8) - 2 wins 2 losses
$(39) - 1 win 3 losses
$(39) - 1 win 3 losses
$(39) - 1 win 3 losses
$(39) - 1 win 3 losses
$(59) - 4 losses
$ 0 - Average Outcome
So on average the SYSTEM doesn't lose or gain any money. The average PERSON (2 wins and 2 losses) loses $(8) after 4 rounds. Best gain is $107, worst loss is $(57).
The gains / losses are mirrored for the rich person. The rich person with more money sees this instead.
$ 59 - 4 wins
$ 39 - 3 wins 1 loss
$ 39 - 3 wins 1 loss
$ 39 - 3 wins 1 loss
$ 39 - 3 wins 1 loss
$ 8 - 2 wins 2 losses
$ 8 - 2 wins 2 losses
$ 8 - 2 wins 2 losses
$ 8 - 2 wins 2 losses
$ 8 - 2 wins 2 losses
$ 8 - 2 wins 2 losses
$(38) - 1 win 3 losses
$(38) - 1 win 3 losses
$(38) - 1 win 3 losses
$(38) - 1 win 3 losses
$(107) - 4 losses
$ 0 Average Outcome
System outcome of zero, but average person (2 wins and 2 losses) sees a small $8 gain. Best gain is $57 and worst loss is $(107)
I'm not actually sure the Yard Sale actually says what the author thinks it's saying...
The rich person betting path seems like a variant of the Martingale System (a betting strategy from 18th century France) where each time you lose you double your bet to recoup your losses. See, each time the rich person in a Yard Sale "loses", their next bet gets bigger and allows them to recoup their loss (since the poor guy has more wealth) - this is what allows their average (2 win 2 loss outcome) winnings to be $8.
The payoff matrix in the Martingale System seems very very similar. The "average" person - with reasonable win / loss percentage - will earn a tiny bit of money, but if you get unlucky your losses can be astronomical, if you go on a big loss streak you're basically betting a million dollars to earn $1 back.
posted by xdvesper at 4:03 PM on December 22, 2022 [1 favorite]
That's what the article was REALLY bad at explaining.
Let's assume a 4 iteration game to make it simple, but this applies for the any n-iterations you like.
There are 16 different outcomes, if you start with $100, listed below.
$107 - 4 wins
$ 38 - 3 wins 1 loss
$ 38 - 3 wins 1 loss
$ 38 - 3 wins 1 loss
$ 38 - 3 wins 1 loss
$ (8) - 2 wins 2 losses
$ (8) - 2 wins 2 losses
$ (8) - 2 wins 2 losses
$ (8) - 2 wins 2 losses
$ (8) - 2 wins 2 losses
$ (8) - 2 wins 2 losses
$(39) - 1 win 3 losses
$(39) - 1 win 3 losses
$(39) - 1 win 3 losses
$(39) - 1 win 3 losses
$(59) - 4 losses
$ 0 - Average Outcome
So on average the SYSTEM doesn't lose or gain any money. The average PERSON (2 wins and 2 losses) loses $(8) after 4 rounds. Best gain is $107, worst loss is $(57).
The gains / losses are mirrored for the rich person. The rich person with more money sees this instead.
$ 59 - 4 wins
$ 39 - 3 wins 1 loss
$ 39 - 3 wins 1 loss
$ 39 - 3 wins 1 loss
$ 39 - 3 wins 1 loss
$ 8 - 2 wins 2 losses
$ 8 - 2 wins 2 losses
$ 8 - 2 wins 2 losses
$ 8 - 2 wins 2 losses
$ 8 - 2 wins 2 losses
$ 8 - 2 wins 2 losses
$(38) - 1 win 3 losses
$(38) - 1 win 3 losses
$(38) - 1 win 3 losses
$(38) - 1 win 3 losses
$(107) - 4 losses
$ 0 Average Outcome
System outcome of zero, but average person (2 wins and 2 losses) sees a small $8 gain. Best gain is $57 and worst loss is $(107)
I'm not actually sure the Yard Sale actually says what the author thinks it's saying...
The rich person betting path seems like a variant of the Martingale System (a betting strategy from 18th century France) where each time you lose you double your bet to recoup your losses. See, each time the rich person in a Yard Sale "loses", their next bet gets bigger and allows them to recoup their loss (since the poor guy has more wealth) - this is what allows their average (2 win 2 loss outcome) winnings to be $8.
The payoff matrix in the Martingale System seems very very similar. The "average" person - with reasonable win / loss percentage - will earn a tiny bit of money, but if you get unlucky your losses can be astronomical, if you go on a big loss streak you're basically betting a million dollars to earn $1 back.
posted by xdvesper at 4:03 PM on December 22, 2022 [1 favorite]
risking a fixed fraction of your wealth on this type of game will almost surely impoverish you in the long term, because the expected geometric growth rate is negative. But in this model players don't play against the house, they play against each other and the total wealth is conserved, so if every player 'almost surely' bankrupts themselves in the long term, where does the money go?
Into the hands of that small and ever-decreasing minority of players whose stake on every round is set not always as a fixed fraction of their own wealth, but increasingly often that of their counterparties.
posted by flabdablet at 10:37 PM on December 22, 2022
Into the hands of that small and ever-decreasing minority of players whose stake on every round is set not always as a fixed fraction of their own wealth, but increasingly often that of their counterparties.
posted by flabdablet at 10:37 PM on December 22, 2022
Tackling inequality from the demand side - "I don't know whether, in some deep sense, we can make people want money less or not. But we can change the legal environment so that the things people do to augment their hordes come with lower payoffs and higher risks. The rich then come to behave as if they want money less."
High top marginal tax rates, or perhaps even a wealth cap. Taxes on firm payouts, higher corporate profits taxes, an excess profit margins tax. Normalizing the nationalization of monopolies. All of these interventions might lead wealthy investors and managers to behave as if they are less greedy.posted by kliuless at 11:26 PM on December 22, 2022 [3 favorites]
Alternatively, we might say the status quo — under which the already rich can endlessly extract ever more funds via ever less competitive firms with little cost or penalty, while as individuals we are locked in competition with one another for essential positional goods — encourages people to behave as if we are more greedy than we in some sense truly are. The fact of inequality, and a legal environment that permits and promotes it, can be understood as a tax on virtue. Let’s tax money instead. Let’s reshape our laws so that endlessly extractive behavior just makes less sense.
Whether the humans in our heart of hearts are generous or greedy, naughty or nice, I leave to you dear reader. Regardless, we can build a world in which people just want to do the things that divide and impoverish us a great deal less. Let’s.
Gary Stevenson, a couple of whose videos I linked above, has repeatedly made the point that since high-value assets will typically return at least 3% of that value yearly as income for their owners (who would not bother owning them otherwise), and since so many of those assets are physical and in-country and therefore not susceptible to being hidden, taxing assets (i.e. wealth) at say 1% per year would be as lucrative and as fair as a 33% income tax and far, far harder to avoid.
As things stand at present, the nominally progressive income taxes typically found in advanced economies really only apply to those who make their income by selling some form of labour. But the further up the income scale we look, the more likely it is that the people making that level of income will be doing so not via sale of labour but via ownership of assets, and the easier it will be for them to disguise that income and have it classified as something else and therefore avoid paying income tax on it.
So in practice the income tax regime is not progressive across all income levels, but is instead humped: most of the tax burden falls on workers with moderately high incomes, not on the super-rich who actually have most of the money.
This is unjust. The principle behind progressive income tax is that those members of any society who have had the most benefit from the way it's organized should also be those who pay the lion's share of what it costs to keep that organization stable, and in fact it's in their direct self-interest to do so; see also: French Revolution.
And most such people do.
It's only a tiny handful of deadbeat bludgers at the very top of the wealth scale who free-ride on the rest of us and refuse to pay their way, and the reason they keep on getting away with not paying their way is that the PR required to stop the rest of us from demanding that they do costs way less than any conceivable principle of natural justice would say is their due.
Wealth taxes are not even novel; taxes on assets are already a matter of common practice. In Australia, for example, local government is funded by council rates, a land value tax. As far as I'm aware there are no good arguments against shifting the mix of State revenue further toward those and away from income taxes. Because income is not something we actually want to disincentivize.
Hoarded dynastic wealth, by way of contrast, fucks everything up and we would benefit from disincentivizing that. Private property that has grown to such a cancerous extent as to cause widespread public deprivation no longer has a moral claim on public protection.
posted by flabdablet at 12:19 AM on December 23, 2022 [1 favorite]
As things stand at present, the nominally progressive income taxes typically found in advanced economies really only apply to those who make their income by selling some form of labour. But the further up the income scale we look, the more likely it is that the people making that level of income will be doing so not via sale of labour but via ownership of assets, and the easier it will be for them to disguise that income and have it classified as something else and therefore avoid paying income tax on it.
So in practice the income tax regime is not progressive across all income levels, but is instead humped: most of the tax burden falls on workers with moderately high incomes, not on the super-rich who actually have most of the money.
This is unjust. The principle behind progressive income tax is that those members of any society who have had the most benefit from the way it's organized should also be those who pay the lion's share of what it costs to keep that organization stable, and in fact it's in their direct self-interest to do so; see also: French Revolution.
And most such people do.
It's only a tiny handful of deadbeat bludgers at the very top of the wealth scale who free-ride on the rest of us and refuse to pay their way, and the reason they keep on getting away with not paying their way is that the PR required to stop the rest of us from demanding that they do costs way less than any conceivable principle of natural justice would say is their due.
Wealth taxes are not even novel; taxes on assets are already a matter of common practice. In Australia, for example, local government is funded by council rates, a land value tax. As far as I'm aware there are no good arguments against shifting the mix of State revenue further toward those and away from income taxes. Because income is not something we actually want to disincentivize.
Hoarded dynastic wealth, by way of contrast, fucks everything up and we would benefit from disincentivizing that. Private property that has grown to such a cancerous extent as to cause widespread public deprivation no longer has a moral claim on public protection.
posted by flabdablet at 12:19 AM on December 23, 2022 [1 favorite]
I believe this is all just a mathematization of noted economist Billie Holiday's statement, "Thems that got shall get, thems that not shall lose. So the Bible says, and it still is news."
posted by clawsoon at 4:56 AM on December 23, 2022 [2 favorites]
posted by clawsoon at 4:56 AM on December 23, 2022 [2 favorites]
Why Rich People Get Richer yt
That video is bad. His basic explanation is that rich people get richer because they own property, or more basically rich people get richer because they were rich in the past. I'm sorry, but that's borderline useless and dumb. It also skips a bunch of steps, ie: how do they come to own all those assets in the first place? Well, they are rich, ergo they buy stuff. Thanks.
No, the answer is actually that if you have as little as $1m in actual wealth, whether that's stock market or a home, it'll spin off about $40-50k in asset growth per year on average. Which is right at the median individual income for the US. Bump that to $2m, and you make more simply by owning your assets that 2 people do from working. At about $3.5m, you surpass the top 10% of households in earnings per year. At $4, you are at top 5%. Why waste your time contemplating $20m, or even $1B when you surpass almost everyone in the US' yearly working income at $5m in assets?
So this tells you that stretching to get on the property ladder (even if it's just your personal home) is an amazing investment strategy. It's basically forced savings for everyone on the lower end of the spectrum. And that's how rich people get rich to begin with. They have a good job, maybe involving taking a risk, and they earn for a while, and they invest in hard assets.
posted by The_Vegetables at 8:46 AM on December 23, 2022
That video is bad. His basic explanation is that rich people get richer because they own property, or more basically rich people get richer because they were rich in the past. I'm sorry, but that's borderline useless and dumb. It also skips a bunch of steps, ie: how do they come to own all those assets in the first place? Well, they are rich, ergo they buy stuff. Thanks.
No, the answer is actually that if you have as little as $1m in actual wealth, whether that's stock market or a home, it'll spin off about $40-50k in asset growth per year on average. Which is right at the median individual income for the US. Bump that to $2m, and you make more simply by owning your assets that 2 people do from working. At about $3.5m, you surpass the top 10% of households in earnings per year. At $4, you are at top 5%. Why waste your time contemplating $20m, or even $1B when you surpass almost everyone in the US' yearly working income at $5m in assets?
So this tells you that stretching to get on the property ladder (even if it's just your personal home) is an amazing investment strategy. It's basically forced savings for everyone on the lower end of the spectrum. And that's how rich people get rich to begin with. They have a good job, maybe involving taking a risk, and they earn for a while, and they invest in hard assets.
posted by The_Vegetables at 8:46 AM on December 23, 2022
how do they come to own all those assets in the first place?
Judicious choice of parents, mostly.
posted by flabdablet at 12:35 PM on December 23, 2022 [3 favorites]
Judicious choice of parents, mostly.
posted by flabdablet at 12:35 PM on December 23, 2022 [3 favorites]
« Older The Spirit answered not, but pointed onward with... | The Fight Over College Athletics Heats Up In South... Newer »
This thread has been archived and is closed to new comments
posted by mmcg at 1:11 PM on December 21, 2022 [2 favorites]