collective action problem

[Metric]

So, some people including Stef spend a lot of time talking about DRO's and other ways that markets can solve traditional problems of the state.  Basically, I think these arguments would be supercharged if they could be used to construct solutions to "collective action problems" in game theory language, since this seems to be a major justification for statism amongst sophisticated statists.  Is there a way to cram a "DRO strategy" (or whatever you want to call it) into game theory language, and show that it can resolve long-standing problems?

For example, I know that Stef and others have claimed that DRO's represent a solution to air quality, which is a textbook collective action problem -- it would be super awesome to see this put into game theoretic language and resolve the paradox (that it is profitable/rational for each individual actor to pollute, even though there is a huge net value gain if everyone cooperates and does not pollute).

Here is an intro to the collective action problem by a guy who does a lot of "intro to game theory" youtubes: 

[Metric]

Just to clarify, I welcome any opinions on this matter -- not only explicit game theory solutions.  :-)

[TheRobin]

well, the problem of any "public good" is first and foremost, it doesn't exist So what kind of solution do you expect?However in regards to game theory, an intersting thing could maybe be done, which statists often seem to ignore when trying to fit their agenda into descriptions of GT. That is if one introduces a third player with the ability to force other players (i.e. the state), then this generates a new action among all former players which is "bribe the state", so the game changes to "at what point is it more profitable to bribe the state then not to bribe the state". Any attempts to create a "checks and balance" also fails for the same reason, as it merely introduces new players and new actions for bribery.So imo, if you want some fun with GT, I'd go for this approach

[MrCapitalism]

 

 even though there is a huge net value gain if everyone cooperates and does not pollute).

 

 

I don't remember where I saw it (somewhere in mises.org), but I'll try to paraphrase the answer until/if I find it.

But the underlying statement in the "tragedy of the commons" type of arguments is that the net benefit of collective action far outweighs the costs acruing to any individual participant. Or put more simply, that there is a profit to be found in solving the problem of the common (in this case air polution). The argument rephrased is that private profit seeking individuals will be unable to exploit a profitable situation, even though there is a stated economic gain in solving the problem.

Tom Woods jokingly give a sort of answer.

 

[Metric]

 

 

 even though there is a huge net value gain if everyone cooperates and does not pollute).

 

 

I don't remember where I saw it (somewhere in mises.org), but I'll try to paraphrase the answer until/if I find it.

But the underlying statement in the "tragedy of the commons" type of arguments is that the net benefit of collective action far outweighs the costs acruing to any individual participant. Or put more simply, that there is a profit to be found in solving the problem of the common (in this case air polution). The argument rephrased is that private profit seeking individuals will be unable to exploit a profitable situation, even though there is a stated economic gain in solving the problem.

Tom Woods jokingly give a sort of answer.

 

 

 

So in terms of game theory, the game has to be generalized in some way to allow for additional plays, as the standard type of game (see link in OP) is artificially restricted (only two moves are possible -- cooperate or not).  This definitely makes sense, as there seems to be massive profit potential to be exploited in some way, and there exists enormous creativity in sophisticated free markets.  The non-excluable, non-rival axiom of the game is the way the narrow range of possibilities are justified -- so one may be able to argue one of the following points a) no such non-excludable, non-rival situation actually exists in real life or b) the non-excludable, non-rival axiom doesn't actually translate into the commonly accepted limitations on the game.

 

[Metric]

 

well, the problem of any "public good" is first and foremost, it doesn't exist So what kind of solution do you expect?


However in regards to game theory, an intersting thing could maybe be done, which statists often seem to ignore when trying to fit their agenda into descriptions of GT. That is if one introduces a third player with the ability to force other players (i.e. the state), then this generates a new action among all former players which is "bribe the state", so the game changes to "at what point is it more profitable to bribe the state then not to bribe the state". Any attempts to create a "checks and balance" also fails for the same reason, as it merely introduces new players and new actions for bribery.

So imo, if you want some fun with GT, I'd go for this approach

 

And this would be argument c) (to be fair it should have been argument a) since you responded first ).  That altough the game may accurately represent a real problem in real life, the "state solution" is actually not really a solution (it is only assumed to be without examination), as it inevitably changes and expands the game in such a way that the problem becomes worse, not better.

 

[ribuck]

For example, I know that Stef and others have claimed that DRO's represent a solution to air quality, which is a textbook collective action problem -- it would be super awesome to see this put into game theoretic language and resolve the paradox (that it is profitable/rational for each individual actor to pollute, even though there is a huge net value gain if everyone cooperates and does not pollute).

I can resolve this paradox rigorously, although I don't know whether I'm using game theory.

Suppose there are 100 people. Each individual who pollutes will gain $10. If no-one pollutes, there is a net gain to their society of $2000. One person runs a DRO, and offers $19 to each of the other 99 people provided that (a) they stop polluting, and (b) the DRO receives the $2000 net gain, but © the agreement is void unless all 99 people sign up.

This doesn't work, because some of the people think "The DRO will never get all 99 people to sign up, so I won't bother". As a result, only 95 people sign up and the net gain isn't realised.

So the following year, the DRO guy tries something more sophisticated. Like a true entrepreneur, he decides to risk some of his own money. He says to the other 99 people: "I will pay you $11. It's yours to keep, with no strings attached, if you (a) stop polluting, and (b) agree that my DRO receives the $2000 net gain if everyone stops polluting. In addition, I will pay you a further $5 if everyone signs up."

This succeeds. If every polluter can gain $10 by polluting or $11 by not polluting, of course they will each agree individually to not pollute. Because everyone prefers $11 to $10, everyone signs up and stops polluting. The DRO guy receives the $2000 net gain. From that, he takes $11 for himself per person (to reimburse himself for the money that he risked by paying everyone $11 up-front with no strings attached), then he pays everyone the extra $5 bonus because everyone signed up, and the remaining $400 is his profit.

We can then consider a society with two DROs who are competing to get people to stop polluting. Naturally they will cut their profit margin, in order to persuade as many people as possible to sign with them. So maybe only $100 of the non-pollution "windfall" will go to the winning DRO as profit. But now, each DRO will have signed up some of the people. Obviously it is now in the economic interest of both DROs to combine their lists, win the $2000 net gain, and split it between them.

This whole apparent paradox is resolved so much more neatly and so much more profitably than the statist solution, where the government bans everyone from polluting, except for those who lobby hard enough and donate enough money to it, and except for some of its own government departments who continue to pollute.

[MrCapitalism]

So in terms of game theory, the game has to be generalized in some way to allow for additional plays, as the standard type of game (see link in OP) is artificially restricted (only two moves are possible -- cooperate or not).  This definitely makes sense, as there seems to be massive profit potential to be exploited in some way, and there exists enormous creativity in sophisticated free markets.  The non-excluable, non-rival axiom of the game is the way the narrow range of possibilities are justified -- so one may be able to argue one of the following points a) no such non-excludable, non-rival situation actually exists in real life or b) the non-excludable, non-rival axiom doesn't actually translate into the commonly accepted limitations on the game.

To put it in terms of the OP video; If any of the 100 individuals will refuse to provide the good any time that the individual cost is greater than 3, then 29,500 units of productivity are lost. This means any entrepeneur who is capable of devising a means of excluding non-contribiters will reap a benefit of up to 29,499-e units of productivty, where e = the cost of implemention of the solution.

 

 

[MrCapitalism]

Here it is!

Hans Hermann Hoppe - Fallacy of Public Goods Theory and the Production of Security.

Pg 28-29 are where he starts to get into it.

 

EDIT: Freedomain Radio is a non-rivalous website which chooses to be non-exclusive and which is successfully privately provided.. a clear violaton of game theory.

[Metric]

 

For example, I know that Stef and others have claimed that DRO's represent a solution to air quality, which is a textbook collective action problem -- it would be super awesome to see this put into game theoretic language and resolve the paradox (that it is profitable/rational for each individual actor to pollute, even though there is a huge net value gain if everyone cooperates and does not pollute).

I can resolve this paradox rigorously, although I don't know whether I'm using game theory.

Suppose there are 100 people. Each individual who pollutes will gain $10. If no-one pollutes, there is a net gain to their society of $2000. One person runs a DRO, and offers $19 to each of the other 99 people provided that (a) they stop polluting, and (b) the DRO receives the $2000 net gain, but © the agreement is void unless all 99 people sign up.

This doesn't work, because some of the people think "The DRO will never get all 99 people to sign up, so I won't bother". As a result, only 95 people sign up and the net gain isn't realised.

So the following year, the DRO guy tries something more sophisticated. Like a true entrepreneur, he decides to risk some of his own money. He says to the other 99 people: "I will pay you $11. It's yours to keep, with no strings attached, if you (a) stop polluting, and (b) agree that my DRO receives the $2000 net gain if everyone stops polluting. In addition, I will pay you a further $5 if everyone signs up."

This succeeds. If every polluter can gain $10 by polluting or $11 by not polluting, of course they will each agree individually to not pollute. Because everyone prefers $11 to $10, everyone signs up and stops polluting. The DRO guy receives the $2000 net gain. From that, he takes $11 for himself per person (to reimburse himself for the money that he risked by paying everyone $11 up-front with no strings attached), then he pays everyone the extra $5 bonus because everyone signed up, and the remaining $400 is his profit.

We can then consider a society with two DROs who are competing to get people to stop polluting. Naturally they will cut their profit margin, in order to persuade as many people as possible to sign with them. So maybe only $100 of the non-pollution "windfall" will go to the winning DRO as profit. But now, each DRO will have signed up some of the people. Obviously it is now in the economic interest of both DROs to combine their lists, win the $2000 net gain, and split it between them.

This whole apparent paradox is resolved so much more neatly and so much more profitably than the statist solution, where the government bans everyone from polluting, except for those who lobby hard enough and donate enough money to it, and except for some of its own government departments who continue to pollute.

 

Very nice.  I'll have a go at modeling this in game theory terms, and also look at the Hoppe article linked by Mr. Capitalism.  If it all goes together nicely, generalizes well, and is an under-appreciated point at mises.org and elsewhere (i.e. Hoppe etc. hasn't already scooped us), I wouldn't mind writing or possibly co-writing an article for mises.org with you about this point.

Incidentally, "argument c" above is probably independently true, and can probably be modeled quite easily.  One basic issue is that the "bribing government" move probably has a relatively fixed cost, while government regulation is presumably a cost that is proportional to the amount of your output.  Thus the bribe move is profitable for "big business" but not for the mom and pop -- this means that "big business" gets a market advantage that thends to destroy mom and pop, and thus the "bribe move" becomes more and more important in the economy, eventually leading to a fascist sort of economy dominated by big business (sound familiar?).  It will be cool to see this work in a little toy game theory model.

 

[Livemike]

Bear in mind that solving a "public goods problem" is an exercise in cooperation, and we happen to belong to most successfully cooperative species in the known universe.  It shouldn't be THAT hard.  Certainly simpler than solving the problem (itself a "public goods/collective action problem") of making the state behave decently.