"Society works best as a whole when individuals are acting entirely within their own self interest."And I closed the post promising to disprove the hypothesis, at least in part, using game theory.
I have to confess that I feel a bit like the Great Gonzo from The Muppet Show, promising an incredible but ridiculous stunt. But I assure everyone, I am dead serious in this endeavor. For the non-geeks out there, I will start from the practical side of things and delve into the math as it becomes necessary for the proof.
In 1968, the ecologist Garrett Hardin discovered an alarming pattern in nature called the Tragedy of the Commons. Simply put, individuals acting purely out of self interest can, and will, deplete resources to the point of extinction. The result is an irrecoverable state. This pattern has been observed, for example, in the fishing industry, where certain species of fish have been depleted to the point where their stocks cannot recover. With non-renewable resources, such as fossil fuels, there is no possibility of recovery at all. Fossil fuels have an additional penalty because they emit greenhouse gases--a second Tragedy of the Commons that will lead to an atmosphere too high in carbon dioxide if not consumed in a guided manner.
OK, now here is the math part for the geeks. It is easy to model the Tragedy of the Commons using game theory. Some of the more common isomorphic games are the Diner's dilemma and the Prisoner's dilemma. Simply put, the optimal solution to the game, if people are behaving out of pure self interest, is for everyone to lose. For those not familiar with game theory terminology, in order for the isomorphism to work, we have to assume that something called a Nash equilibrium represents the state of all solutions if everyone is behaving rationally and wants to maximize their own gain (payoff). The way to arrive at a Nash equilibrium is by eliminating all strongly dominated strategies. In other words, if the most you can gain by adopting a particular strategy is lower than the least you would gain by adopting a different strategy, you discard that strategy. As you keep iterating through, assuming that each player is doing the same thing, you eventually arrive at the minimum set of strategies that would be taken in a "rational" game. Once you are there, the strategies that provide the maximum payoff for each player with full knowledge of the other players' possible choices are the Nash equilibria.
Obviously, simply because you can model X as Y, it doesn't mean that it is necessarily so. However, game theory has proven itself to be quite accurate in describing many other naturally occurring phenomena, and there is no reason to assume that a Prisoner's dilemma is only a fictitious thing. I can't, for example, prove that the Hand of God won't suddenly appear and give us more oil just as we run out, or that He won't show up with a giant squeegee when the atmosphere is looking a little grimy. I only put it outside of the realm of probability from what we can observe.
For people who think that the threat of environmental damage is overblown or that the consumption of natural resources is not a particularly big issue, I have another example of the Prisoner's dilemma that is a bit more familiar, at least in my own personal experience. As many people may know, I dislike driving. I call this one the Commuter's dilemma.
Mr. Peter Gibbons is a programmer at Initech. Every morning he has to commute to work. Each day is a grueling challenge, as he finds that traffic reaches a bottleneck where his drive slows to a crawl. Suppose that he is on a two lane highway, with lanes A (the fast lane) and B (the slow lane). Mr. Gibbons notices that while he is stuck in lane B, the traffic appears to be moving in lane A. Looking for an opportunity, he swerves out from B into A. In a short amount of time, lane A, too, slows down, while lane B starts moving. So Mr. Gibbons waits for the opportunity, and goes back into lane B. The result is a very frustrated Mr. Gibbons! But, nevertheless, we realize that he is maximizing his potential by switching lanes when he can. Let's say he has saved 10 seconds in his overall commute by adopting the changing lanes strategy. Not bad, Mr. Gibbons! Only, now we discover something. Each time Mr. Gibbons switches lanes, the person behind him, Mr. Michael Bolton (not the singer) is slowed down by 10 seconds in his commute. He, too, can adopt the same lane-changing strategy, but if he does so, both he and Mr. Gibbons will be slowed down by 15 seconds, making it a net loss for the both of them. If Mr. Gibbons stayed in his lane but Mr. Bolton decided to adopt the lane-changing strategy, it is a net gain for Mr. Bolton, but Mr. Gibbons is slowed down by 10 seconds! And yet, let's assume that neither is penalized or rewarded by any amount of time (0) if neither one changes lanes through their commute.
Very good, now that I have thoroughly depressed everyone by showing how self interest is both hopeless and unavoidable at the same time, in my next post I will show how nothing is really hopeless, enlightenment and progress are really possible, and even how the biggest problems can be solved once we get rid of a slight amount of defeatism, denial, and negativity.
This isn't a problem with enlightened self-interest -- it is a problem of lack of information, which happens in any system.
ReplyDeleteThe Tragedy of the Commons is easily solved -- note that it is in nobody's self-interest to keep the commons usable. So, make it that way. You either have someone own the land, and therefore they lease out portions, or you have the community appoint a guardian (effectively having the community 'own' the land and protect it). So the solution is to actually make people have an interest to protect.
The problem with an attack on capitalism/libertarianism/etc is that people seem to have the fantasy of a philosopher king. A Jed Bartlett running the country with perfect knowledge and all problems scripted to happy endings, if only people would just do what he knows is right. That's not how people are. That's not how the people are that are elected.
That leaves you, Mike, to choose who is to become the supreme ruler, and just allow them to create whatever edicts they need to do in their enlightened state of knowledge to tell who needs to do what.
Didn't you play "Paranoia"? ;)
Tal: it's like a conversation I've always envisioned:
ReplyDeletePerson A: let's get rid of the government!
Person B: great, let's!
A and B: now there's no government!
Person A robs Person B.
Person B: if only we had some sort of...government to keep people from doing that!
Tal: also, the Tragedy of the Commons is still a problem, even with perfect information. Nash equilibria assume perfect information, hence the ability to pick strategies based on eliminating the strongly dominated ones.
ReplyDeleteWhat I keep goading about, I'll get to in Part Three. You are assuming long term access to resources as a different payoff. That isn't included in the game, and it certainly isn't included in the game we call the pure free market!
Long tern access to resources is exactly the game, and it is exactly the game called the free market.
ReplyDeleteNow, I'm not a pure-Libertarian by any stretch of the imagination. I fully value regulations, and definitely believe that government provides an exceptionally valuable service. The problem becomes, in my mind, when we stop believing that individuals can choose to do what is correct, and must be coerced into action.
Examples: banning trans-fats in food. Banning drugs. Drinking ages. Smoking bans. Just because any of these activities "cost" society, that's okay with me, since I'm paying for freedom.
On the contrary, long term access to resources is not the game you get with free market capitalism. Not in every case, obviously, but in enough cases for it to do permanent, untold damage. Science has told us, for instance, that cap and trade is a solution to carbon emissions; unfortunately, it requires both external cooperation from every country and honest enforcement from within--both Russia and China have reasons to "cheat" in order to better their economies at the expense of the rest of the world; the U.S. recognizes this as a strongly dominated strategy and rejects it.
ReplyDeleteThe examples you give are very good ones, all worthy of individual discussions IMO, and in general I don't disagree with you--but I have to argue that both prohibition and anarchy are suboptimal and have problems. You can't always do *nothing*, and you can't deprive everyone of choice. What you can do is, as my friend Mordicai has suggested separately in one of his own blog posts:
http://mordicai.livejournal.com/1827382.html
...to use both "the carrot & the stick" as adaptive tools to guide society.
Smoking bans are a good example to use because it is the one you give that doesn't match the others (at least in the context of today's terms), because as a member of society, I don't want to have to be involuntarily subjected to somebody else's poison just because I favor freedom, shortening my life without my consent. On the other hand, I don't want to deprive others of their right to choose. Still, if the substance is addictive and the cost to society is bad enough (web search: heart disease), arguably as an individual I don't want to have to subsidize it (collectively driving up health care costs for everyone). It is a balance that I would lean heavily towards the "prohibition" side, but I would say that the correct remedy is a tax that offsets the additional cost--preferably one that covers the additional costs of insurance and funds substance treatment programs.
...continued...
ReplyDelete"The problem becomes, in my mind, when we stop believing that individuals can choose to do what is correct, and must be coerced into action."
Sometimes this is EXACTLY the case, though. We saw what Wall Street did--hell, we saw it coming a mile away, and yet, the government did nothing. It wasn't until after the damage was done that somebody said, "maybe we ought to reform banking." But that's also a case of closing the barn door after all the animals have already left. Before derivatives, it was Enron and Sarbanes-Oxley. Before that, it was the Savings and Loans scandal, etc., etc. At some point, the argument starts to look good that maybe you should start with the exact assumption--that people won't do what's right and must be coerced--and come up with a more strategic, or, as you put it, centralized solution to the ongoing problem.
Or maybe we can just take the tactic that everyone who suffered during the recession was collateral damage and "deserved it" (whatever that means) while CEOs "earned" their money by being good alchemists and figuring out how to cheat everyone else. I know this is a straw man, but some people are saying exactly this...and they are pushing Tea on everyone...
Is there a term for "greatest total pay-out?" Nash shoots for the highest for one player; is there a game theory term for the greatest value for ANYBODY?
ReplyDeleteAlso, did you read about this (from Wikipedia):
"...a team from Southampton University in England (led by Professor Nicholas Jennings [2] and consisting of Rajdeep Dash, Sarvapali Ramchurn, Alex Rogers, Perukrishnen Vytelingum) introduced a new strategy at the 20th-anniversary iterated prisoner's dilemma competition, which proved to be more successful than tit for tat. This strategy relied on cooperation between programs to achieve the highest number of points for a single program. The University submitted 60 programs to the competition, which were designed to recognize each other through a series of five to ten moves at the start. Once this recognition was made, one program would always cooperate and the other would always defect, assuring the maximum number of points for the defector. If the program realized that it was playing a non-Southampton player, it would continuously defect in an attempt to minimize the score of the competing program. As a result,[6] this strategy ended up taking the top three positions in the competition, as well as a number of positions towards the bottom."
Does this sound like no-bid government contracts? Or hey, any piece of unregulated capitalism with strong evangelical arms (re: advertising, publicity)? 'cause it does to me.
I find it interesting that you point at enforced cap-and-trade in and in the same sentence point out how countries will cheat because it benefits them. We are very unlikely to go to war to stop carbon emissions. And Carbon Trading therefore has had a very interesting effect in Europe: http://arstechnica.com/science/news/2010/03/europe-outsourcing-co2-emissions-to-developing-economies.ars . Europe has had massive incentives to move manufacturing out to areas that are cheaper due to less strict Carbon emissions, and has likely ended up with a net-world increase in CO2.
ReplyDeleteAnd that's the problem with any regulations. Congress passes wonderful concepts, "Banks must have much more capital!", isn't specific about what that means, the rules shake out and in the mean time, people aren't getting money for small business loans. Then, after all the rules settle, if there's still a way to get some more profit, that happens.
And Government did far from nothing -- many would say how Government, through rules and regulations, actually stoked the fires:
" In 1992, Congress gave a new affordable housing "mission" to Fannie and Freddie, and authorized the Department of Housing and Urban Development to define its scope through regulations. Shortly thereafter, Fannie Mae, under Chairman Jim Johnson, made its first "trillion-dollar commitment" to increase financing for affordable housing. What this meant for the quality of the mortgages that Fannie--and later Freddie--would buy has not become clear until now.
On a parallel track was the Community Reinvestment Act. New CRA regulations in 1995 required banks to demonstrate that they were making mortgage loans to underserved communities, which inevitably included borrowers whose credit standing did not qualify them for a conventional mortgage loan." ( http://www.forbes.com/2009/02/13/housing-bubble-subprime-opinions-contributors_0216_peter_wallison_edward_pinto.html )
Now, I am not saying that "poor people are to blame". However, the theme of not-so-great loans combined with an incentive to push more money into housing certainly did not help matters at all -- and nobody was clamoring for tighter restriction. If anything, people were demanding oversight to get more loans out to underrepresented people. This is what Government Oversight was able to do.
Every recession prompts a call towards, "if only the government could run everything, these things wouldn't happen." Booms tend to see the sentiment go the other direction. Wars tend to demonstrate why we _don't_ want governments running everything.
Tal, I am enjoying the discussion and your contributions. I am allowing myself to get sidetracked a bit, though, which is no fault of yours!
ReplyDeleteI do agree with the caution of falling into the trap of being reactive and saying "if the government could run everything..." But, to reiterate, this is not what I am saying. I am only pointing out that some things DO need to be centralized, and that the appearance of a Prisoner's dilemma is a good indicator of where some form of central role needs to be played.
The article about carbon emissions in Europe is rather interesting, and, if anything, suggests that social injustice and poverty are bigger human problems than carbon emissions and may be prerequisites to a solution.
In response to "many would say how Government, through rules and regulations, actually stoked the fires...", all I would say is that I hope you aren't one of the "many" who would buy this argument. There is a big difference between bad local legislation and broad macroeconomic policy. It's a bit like performing an external exam instead of an autopsy and concluding that it was the patient's conjunctivitis that did him in. It was, in fact, deregulation that unequivocally precipitated the credit crisis: allowing the formulation of SPVs (special purpose vehicles) for credit derivatives, no oversight into how they are valuated, traded (over the counter, without a public exchange), no accountability in balance sheets, and then selectively bailing them out, at taxpayer expense, when the whole system went kablooie (with nepotism and cronyism being all too transparent in the process).
Mordicai: that is kind of awesome, and shows how monopoly / oligopoly / gentleman's agreements can successfully game the system and break it.
ReplyDeleteMordicai: "Is there a term for "greatest total pay-out?" Nash shoots for the highest for one player; is there a game theory term for the greatest value for ANYBODY?"
ReplyDeleteThere isn't, but collaborative (instead of competitive) games in single payoff terms are trivial to compute--which is why people probably found them uninteresting. It kind of makes the argument in favor of cooperation as a solution to situations like the Prisoner's dilemma, and fits rather well into my earlier post about people tending to be more benevolent and collaborative in nature than we give them credit for ("The Little Old Lady Experiment").