The Cold War can be viewed as a massive Prisoner’s Dilemma, probably the largest one that’s ever existed. If either the USA or USSR had not built nuclear weapons (“cooperated”) and the other had, they would have been at a massive disadvantage and probably gotten themselves steamrolled. They both had to do the same thing, and so paid the military and economic price of building tons and tons of nuclear weapons. If neither side had ever built weapons, all that money could have gone to more productive efforts, so there was a very real price paid. Several attempts, notably the SALT talks, were made to institute cooperation between the two nations, but trust never existed between them, and true disarming could never take place without that level of trust.
You can see this same sort of thing going on all the time in political campaigns, notably when it comes to “dirty campaigning”. Both sides decry it, but it works. Campaigns which unilateral “disarm” often get crushed. If both sides cooperated, the country and governance would have greater trust from the public and we would be in better shape. But the trust required to truly cooperate rarely exists, and so both sides defect and play dirty. And they pay a price in “political capital” – distrust from the public and loss of ability to get things done when elected.
More recently and relevantly, I am of the opinion that economic downturns can be regarded, in some sense, as a multiplayer Prisoner’s Dilemma. Here the players are the various corporations that make up our economy, and defecting is cutting spending or engaging in layoffs. Now, this is not really a straightforward application of the Prisoner’s Dilemma due to the number of players and the fact that the players are not equal and the penalties they pay are not equal. But there are definitely some similarities.
Companies rely on other companies to stay in business. Nevertheless, any given company’s actions are generally taken with an eye exclusively to that company’s interests, and not based on the economy as a whole. But if no company were to defect – slashing expenditures or workforce – there would, almost by definition, be no downturn. (I grant that the downturn we are in right now had a definite cause – the credit crunch – and so can not be viewed as a typical example.) The collective reward for mass cooperation is great, but the reward for a defection (layoffs etc) is greater. Without external action, it can become a slippery slope.
Speaking of the credit crunch, though, the lack of transparency into the financial instruments involved was a catastrophe in the making. Nobody knew what they were buying; they trusted in the market price as an accurate indicator of value. But it can be argued that this lack of transparency itself was caused by a perceived Prisoner’s Dilemma: CDO issuers who provided transparency would be at a disadvantage to those who did not. If all had been transparent, that disadvantage would have been lessened (but there still would be a small penalty in that CDOs would not appear as attractive compared to bonds and stocks as they did in reality).
And this, really, is what I think one of the key roles of government should be: to break the Prisoner’s Dilemma. Regulations can effectively remove the option for defection altogether, thus ensuring mass cooperation amongst all the parties. If CDOs had been regulated with adequate transparency, they wouldn’t have flown so high nor crashed so hard.
But the government doesn’t always have to manipulate the game – in times such as these, the government can also act as a player in the national economy. By acting as an always-cooperating player, the government can encourage some cooperation among other players (companies) and start to pick things up again. Stimulate the economy, as it were. This is the sort of action I see Obama’s stimulus plan accomplishing.
Can everything be viewed through the lens of the Prisoner’s Dilemma? Of course not. Do some of my examples stretch it a bit? Well, yeah. But I maintain it is a useful tool to use to examine some actions, and why government is needed from time to time to “interfere” in the free markets.
… illustration of ‘the problem of collective action’. I’d nitpick, however that this is the problem you’re really interested in from a policy point of view. Strictly speaking, the PD is a particular subset of the collective action problem. In particular, it has one component that was left unstated above. Each prisoner has limited information. More specifically, each isn’t informed as to what the other prisoner is going to do. This lack of information is key in the applicability of the model to the examples you cite. I’m not sure that the PD is actually applicable in most policy cases, because there is usually some information (beyond the game ‘rules’) out there that actors can use in formulating strategy.
<
p>I was going to chime about how when mortgage brokers would try to sell a particular type of loan with particular credit risk to investment banks (particular because the people they were selling them to packaged them into loans of ‘like’ properties), they would pressure each until one caved. Once one caved they could use that fact to add pressure on the others and before you knew it all the banks were buying what was previously too risky to sell. The buyers are operating on game theory that they don’t want to be left behind while other banks improve (heh) their market position.
<
p>The example popped into my head, but then I remembered about the information part. Their behavior is based on more information than the prisoners have.
has a decent amount of explanatory power when applied to the Israel-Palestinian conflict. The sides are clearly defined, though hardly monolithic. There’s an overriding fear that one side’s concessions will lead to the other advantage. What’s needed is a win-win situation.
<
p>Mark