Posts Tagged ‘prisoner’s dilemma’

John Cassidy at The New Yorker has a post at his new blog, Rational Irrationality, defending his position from an article last week that the financial crisis was caused by individuals making completely rational decisions that, when aggregated at the collective level, turned out to be completely irrational. In the original article, Cassidy explains that the Prisoner’s Dilemma is at the root of the financial crisis, and he’s certainly right about that. Cassidy’s explanation of the basic dynamics of the Prisoner’s Dilemma and their application to business and finance is as good as you’ll find, so I’ll quote it in full here. (Note: Charles Prince was the Citigroup CEO when Citigroup got into the business of CDOs.)

Because financial markets consist of individuals who react to what others are doing, the theories of free-market economics are often less illuminating than the Prisoner’s Dilemma, an analysis of strategic behavior that game theorists associated with the RAND Corporation developed during the early nineteen-fifties. Much of the work done at RAND was initially applied to the logic of nuclear warfare, but it has proved extremely useful in understanding another explosion-prone arena: Wall Street.
Imagine that you and another armed man have been arrested and charged with jointly carrying out a robbery. The two of you are being held and questioned separately, with no means of communicating. You know that, if you both confess, each of you will get ten years in jail, whereas if you both deny the crime you will be charged only with the lesser offense of gun possession, which carries a sentence of just three years in jail. The best scenario for you is if you confess and your partner doesn’t: you’ll be rewarded for your betrayal by being released, and he’ll get a sentence of fifteen years. The worst scenario, accordingly, is if you keep quiet and he confesses.
What should you do? The optimal joint result would require the two of you to keep quiet, so that you both got a light sentence, amounting to a combined six years of jail time. Any other strategy means more collective jail time. But you know that you’re risking the maximum penalty if you keep quiet, because your partner could seize a chance for freedom and betray you. And you know that your partner is bound to be making the same calculation. Hence, the rational strategy, for both of you, is to confess, and serve ten years in jail. In the language of game theory, confessing is a “dominant strategy,” even though it leads to a disastrous outcome.
In a situation like this, what I do affects your welfare; what you do affects mine. The same applies in business…. If Merrill Lynch sets up a hedge fund to invest in collateralized debt obligations, or some other shiny new kind of security, Morgan Stanley will feel obliged to launch a similar fund to keep its wealthy clients from defecting. A hedge fund that eschews an overinflated sector can lag behind its rivals, and lose its major clients. So you can go bust by avoiding a bubble. As Charles Prince and others discovered, there’s no good way out of this dilemma. Attempts to act responsibly and achieve a coöperative solution cannot be sustained, because they leave you vulnerable to exploitation by others. If Citigroup had sat out the credit boom while its rivals made huge profits, Prince would probably have been out of a job earlier. The same goes for individual traders at Wall Street firms. If a trader has one bad quarter, perhaps because he refused to participate in a bubble, the results can be career-threatening.

Following the argument in this last paragraph, Cassidy argues that the situation should be dubbed “rational irrationality,” as the rational behaviors of individual traders or firms leads to irrationality at the collective level. In other words, what’s good for all these individuals ends up being bad for the collective of which they’re a part. In this case, that collective happens to include the rest of us, too.

As Cassidy notes in his follow-up blog post, Barry Ritzholtz disagrees that the individual behavior Cassidy describes is rational. Quoting Cassidy:

Ritholtz writes: “ ‘Rational Irrationality’ asks us to ignore the repercussions of our behaviors. We can rationalize short term gains at the expense of long term losses, because we need to obtain quarterly profits regardless. Apparently, when it bankrupts the company, only then with the benefit of hindsight can we see what went wrong. I am terribly sorry, but that is precisely the sort of thinking that led to the crisis in the first place. Making loans to people who cannot pay them back is not rational when its profitable—its NEVER rational.”

As I mentioned, Cassidy defends himself against Ritzholtz’s argument, and at the risk of having this post be overly long, I’d like to quote the defense at length.

From a macro point of view, Ritholtz is right. Directing capital to people or firms who can’t afford to service it amounts to misallocating resources, and it is a form of market failure. But from a micro point of view, things are very different. Most of the people issuing sub-prime loans were loan officers at mortgage lending firms, such as Ameriquest and New Century, who were paid on commission. Neither they nor their bosses had much interest in whether the borrowers could repay the loans, and for good reason. They were intending to sell them on to Wall Street firms for securitization. If the default rates turned out to be higher than expected, it was the purchasers of sub-prime securities that stood to lose out, not the loan originators.
Now, this was certainly myopic thinking, but it wasn’t irrational. During the good years, some of these brokers racked up seven-figure commissions and bonuses. (See, for example, the excellent account of subprime and Alt-A lending in “Chain of Blame,” by Paul Muolo and Mathew Padila.) To be sure, some (many) of them ended up losing their jobs when the bubble burst, but that doesn’t invalidate the point: pushing sub-prime mortgages onto folks that couldn’t afford them was a lucrative business, and the people who were involved in it it were simply reacting to the price signals that the market was providing to them.
Ritholtz focuses on the role of the Wall Street firms. He writes: “On a risk adjusted basis, the behaviors of Citi, Bear, Lehman, New Century and others was hardly rational. Call it whatever you want, but do not forget this simple fact: It was the sort of narrow, risk-ignoring thinking that is ALWAYS rewarded in the short term, and ALWAYS punished in the long term.”
Sadly, this isn’t true either. During a bubble, the risks are rarely evident. If they were evident, asset prices would adjust and the bubble would pop. From the point of view of a trader, or a CEO, the bigger risk is missing out on the easy money that is to be made. In such an environment, the rational thing to do is surf the bubble.

After reading through this argument, it becomes clear that, at root, Cassidy and Ritzholtz disagree about what the word “rational” means. It’s here that we should step back. Both Cassidy and Ritzholtz think that, for any individual, “rational” means whatever is in that individual’s interests. It’s just that they apply different time frames to the interest. Cassidy thinks that short-term interests often outweigh long-term interests because a person looking at the long term to determine his short-term actions may guarantee that he’s not around for the long term. Ritzholtz, on the other hand, thinks that long-term interests can be seen, at least if one has a sense of history, and must be taken into account in the short term. Hence they disagree about what’s “rational”.

I’d like to suggest that they’re both missing part of the picture. They both assume that to declare something “rational” is to declare it morally good. Because something is rational, however, does not mean it is good. To determine whether a practice — selling CDOs, say — is good, I must have a set of criteria about what constitutes goodness. I may then use rational thought to determine whether a particular practice meets those criteria. But if the criteria are flawed, the practice may be rational — i.e. it may follow logically from or be consistent with the criteria — but it will not be good.

Rather than stretch your post-length tolerance any further, I’ll turn to this point in my next post. If you’d like a hint of what’s to come, you might like to read this op-ed from David Brooks at The New York Times earlier this week.

Read Full Post »

At Salon’s How the World Works, Andrew Leonard writes about the high rates of mortgage delinquency and foreclosure in the US, which are reaching historically unprecedented levels. To explain this, Leonard looks back a few years to when the housing crisis first broke in the US.

From the moment politicians began proposing measures to ameliorate the impact of the housing bust on the U.S. economy and individual Americans, critics responded with a strong dose of moral outrage. Why should speculators, flippers, and deadbeat “losers” get help from the government? They screwed up, and they should pay the price. Let them drown in their own option-ARM bile!

(ARM would be adjustable-rate mortgage. Here is Wikipedia’s explanation, for what it’s worth.)

These critics that Leonard summarizes make the seemingly straightforward point that people should have to deal with the consequences of their mistakes. This attitude would seem to be in line with notions of justice. As Baha’u’llah writes in the Lawh-i-Maqsud,

Justice hath a mighty force at its command. It is none other than reward and punishment for the deeds of men. By the power of this force the tabernacle of order is established throughout the world, causing the wicked to restrain their natures for fear of punishment.

So, in order to have justice, we must punish the wicked and reward the good. Justice thus established also trains and gives life to the world and provides the foundation of its stability and order. Baha’u’llah notes in the same tablet:

The structure of world stability and order hath been reared upon, and will continue to be sustained by, the twin pillars of reward and punishment.

and in the eighth of the Ishraqat:

That which traineth the world is Justice, for it is upheld by two pillars, reward and punishment. These two pillars are the sources of life to the world.

So, the critics that Leonard cites seem to align with Baha’u’llah’s call to justice.

As Leonard notes, though, what seems a matter of justice is not so straightforward in it consequences. Punishing those people who acted badly (or at least unwisely) by letting them fail in their mortgages has larger consequences for the economy. Leonard again:

The sensible response to this blast of anger was to observe that society as a whole would pay a cost for mistakes made by the imprudent, and like it or not, for the benefit of the general public, we need to do what we can to cushion the impact inflicted on all of us by the irresponsible.

Without saying it explicitly, Leonard notes that the structure of this problem is very similar to that of a multi-party prisoner’s dilemma, in which the bad actions of a few affect many beyond them, even those who have acted responsibly. (Unfortunately, the wikipedia article doesn’t have much on the multi-party version. This page explains it reasonably well. If you want more depth, look to Thomas Schelling, who showed up in the Freakonomics blog here.)

When we see borrowers begin to fall behind on prime fixed-rate 30-year-mortgages because of job loss or other recession-fallout, it’s not so easy to self-satisfyingly blame them for their own predicament. They are collateral damage, and as they lose their homes, further depressing home prices and crimping economic growth, the gyre continues to widen. If there’s any lesson from this mess, it’s that more aggressive government help is necessary when the economy gets hit by a falling anvil.

Because of the dynamic in which irresponsible actions affect even those who have acted responsibly, Leonard advocates government intervention, even if it aids the irresponsible actors. Surely the just should not suffer the punishment of the unjust. To put it in terms of game theory, Leonard wants an outside power to tilt the rules of the game in favor of not hurting cooperators. That seems reasonable, too, falling within the ambit of both justice and wisdom, a balance which Baha’u’llah also extols:

Take heed, O concourse of the rulers of the world! There is no force on earth that can equal in its conquering power the force of justice and wisdom. I, verily, affirm that there is not, and hath never been, a host more mighty than that of justice and wisdom. Blessed is the king who marcheth with the ensign of wisdom unfurled before him, and the battalions of justice massed in his rear.

So for regulators, Leonard’s advice is well-grounded. At a more fundamental level, though, the people of the earth need to be alive to the dynamics of collective action as embedded in the multi-party prisoner’s dilemma because so many of our current global problems, from the financial crisis to climate change, manifest that basic structure.

At the level of the individual, we need to begin to heed Baha’u’llah’s call in the Hidden Words, Persian, #5:

O SON OF DUST! Verily I say unto thee: Of all men the most negligent is he that disputeth idly and seeketh to advance himself over his brother. Say, O brethren! Let deeds, not words, be your adorning.

Read Full Post »