The Game [Theory] is Rigged

Jimmy McNulty, The Wire
Image via Wikipedia

Thus story thus far: Andrew Sprung criticizes political scientist/political science blogger Jonathan Bernstein‘s “joyous cynicism” regarding the political process. Or to put it another way:

No one, I’m sure, would quarrel with Bernstein for highlighting the possibility that a party leader in McConnell’s position might enhance his party’s interest by putting up resistance to ratification of  this treaty even if he personally believed the treaty to advance U.S. interests — at least, if he knew that the resistance would ultimately fail.

Where Bernstein (judging from his blog’s Comments section) does disturb many readers — me included — is in his suggestion that it is politicians’ right, indeed their duty, to be guided entirely by such calculations. He argues, in effect, that the law of political survival is a necessary, natural, sufficient and therefore desirable prime mover of politicians’ words and actions.

Fellow political scientist/political science blogger Seth Masket replies:

I don’t want to speak for Jon here (I’m sure he’ll have a good post along these lines up shortly), but my response to this is as follows: I don’t celebrate this system. But to complain that politicians will be guided by political calculation is like complaining that businesspeople will be guided by profit maximization or that athletes are too obsessed with winning. It’s not a character flaw; it’s their line of work. Indeed, hoping for politicians who are untethered from political calculations is not only naïve, but sometimes quite dangerous.

But of course, Sprung never once wishes aloud for “politicians who are untethered from political calculations.” His disagreement with Bernstein (and by extension, Masket) is not over whether politicians should make political calculations, but over how they should determine the ends to which those calculations are directed. Is a good legislator one who makes his political calculations based primarily on (A) moral concerns, or (B) rational self-interest? Bernstein and Masket seem to prefer Option B, and this is exactly what Sprung finds so cynical.

Maybe the features of Option B that so disturb Sprung would be more apparent to Bernstein and Masket if we looked at how other sectors of society function on Planet Option B. You guys know what that means: it’s super happy fun thought experiment imagination time!

Consider the policeman. Not just any policeman. I’m talking about Policeman McNulty. What is Policeman McNulty’s job? Most people would argue that it is to maintain order, uphold the law, and keep the innocent residents of his jurisdiction (the sleepy little hamlet of New Hamsterdam) safe. As long as the NHPD rewards officers based solely on how well they fulfill those responsibilities, and does so with perfect efficiency, all is well.

But one day, mean old Commissioner Burrell decides to institute an arrest quota. Policeman McNulty now needs to bust a certain number of people per week in order to keep his job. McNulty realizes that unless he dramatically steps up his weekly arrest rate, his job is in grave danger.

Now, McNulty wants to be good po-lice. But what does that mean, exactly? In the New Hamsterdam of Planet B, it would mean responding to the incentives provided by Burrell, and boosting his arrest rate by any means necessary. This, Option B tells us, is not just part of what it means to be good po-lice: it supersedes the traditional role of the police officer, because rational self-interest dictates that it would.

But if McNulty starts cuffing jaywalkers and planting crack on people who deserve little more than speeding tickets, are we really going to argue that he is good po-lice? Or would we instead argue that he should ignore rational self-interest and fight for principles above the network of punishments and incentives built into New Hamsterdam’s law enforcement bureaucracy?

When it comes to the NHPD, Worldview B seems not just wrong, but deeply, deeply cynical. Furthermore, we have little reason to believe any different when it comes to the real world of politics.

That said, I don’t think Bernstein and Masket accept the premises of Worldview B out of conscious cynicism. Rather, I think career political scientists are likely predisposed to see a certain beauty and elegance in the system they observe that can all too easily be conflated with a sort of ethical purity. Just as Einstein saw God in physics, the biologist sees it in biology, the economist sees it in economics, and the political scientist sees it in political science. To not find some kind of magic in the subject to which one dedicates a lifetime of scholarship — that would be cynical.

But a little bit of perspective shows us that beauty in game theory is not the same thing as moral worth. We need to hold our elected officials to a higher standard than mere responsiveness to incentives. The fact that they might rarely meet that standard does not make us naïve, so long as we acknowledge those failures and their causes.

Enhanced by Zemanta
Advertisements

7 Responses

  1. Planet B exists and is named Earth. Planet A exists only in your imagination. Game theory is meant to explain reality not how reality should be. I don’t know the authors you’re citing in this article and I don’t agree with them if they think that we should be happy that people are rationally self-interested. However, game theory provides many important insights and has high predictive power. If the people and chief or commissioner studied the theory they wouldn’t place a quota on arrests because analysis of the resulting incentives would illuminate an outcome opposite to the intended outcome.

    • Maybe I wasn’t clear enough: I don’t think there’s anything wrong with using game theory as a predictive model for human behavior. What I’m saying is that it can’t do very much to help us make moral judgments about that behavior.

    • Actually, let me put it another way: If you subscribe to Option B, then you’re not just arguing that people are chiefly governed by rational self-interest. In effect, you’re saying that moral goodness is identical to rational self-interest.

      • Not identical, determined by

        What is moral can be defined in may ways, utilitarians, libertarians, idealists, etc all of their own ideas.

        What I’m saying is that option B is what we’ve got and we need to work within Option B to achieve what we want in option A.

        And in response to your first comment: Game theory is often the best way to determine what is moral or immoral because incentive analysis and strategic moves often reveal more about an individuals intentions than their statements. If the outcomes of different strategies can be determined in advance and a decision maker chooses a strategy that produces an outcome inferior to one that may be produced by another strategy than that person may be immoral.

        For example, if an environmentalist like myself intervenes to prevent the slaughter of harmless whales in international waters it may seem as though I am acting in line with my morals and intentions. However, if the data shows that my intervention somehow hurts the whales by disrupting the ecosystem (as in produces whale over population) and many more die than what the whaling ships would have harvested and I intervene anyway, I’m immoral by my own standards.

      • Game theory can determine consequences, but it’s value neutral. We determine what is moral or immoral about those consequences.

        The idea that we need to “work within Option B to achieve what we want in Option A” makes no sense. Either the end goal is our own self-interest or not, no? If you acknowledge some higher goal beyond your own self-interest, then you’re arguing for Option A.

  2. Obviously, McNulty preferred Worldview A, which resulted in him fabricating evidence and serial killers for the Greater Good and getting kicked off the force. Yes, his heart was in the right place, but surely the dangers of pursuing this path are obvious.

    • But Season 5 and talk about good intention aside, what kind of police officer do we want more of on the streets? In the thought experiment above, is McNulty A a better police officer or McNulty B?

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: