Sunday, August 29, 2010

On aging, conservatism, and experimental economics

So, it is standard conventional wisdom that people are liberal when they're young, and conservative when they're old. To the extent that we interpret "liberal" as "eager for change" and "conservative" as "against change," this trajectory is only natural. Especially in the modern world, where things are changing all the time, it may simply come down to a difference in experience: you're less likely to pine for the way the world was thirty years ago if you weren't alive thirty years ago.

But what I am really interested in here is the apparent trend where people become more conservative with respect to economic policies. In this context, the argument about familiarity does not seem to hold. In the United States, the government's economic policies have been trending more conservative for decades, and the familiarity argument would predict that older people should be, on average, more liberal. However, there is a different aspect of familiarity that may be relevant, as it pertains to our beliefs about human nature.

A key aspect of the economic debate between liberals and conservatives is a difference in the assumptions they make about how people will behave when left to their own devices. If you will forgive me for painting complex things with a simple brush, the cartoon versions of these are something like this. Conservatives believe that people are inherently self-serving and lazy, and will work hard only if they are given tangible incentive to do so. From this perspective, progressive tax structures and government programs like welfare and social security are problematic because they take away the incentive to earn money. Liberals, by contrast, believe that everyone is trying hard, that inequality comes largely from societal structures that are beyond individuals' control, and that people should not be punished for the inherent unfairness of society (except maybe those at the top of the pay scale, who have benefitted most from those inequalities).

Now, the most obvious difference between young people and old people is that old people have a lot more experience with other people than young people do. That is, we tend to start out with positive views about human nature, but over time we interact with more and more people, they disappoint us, and we become progressively more cynical. Many conservatives see this as evidence in favor of their position: we start naive, and become conservative when we learn what people are actually like. However, I want to suggest a different explanation, having to do with asymmetries in how we perceive positive and negative deviations from our expectation. Intuitively, this comes down to the fact that we notice whenever we get stopped by a red light, but often don't notice when we hit a green light. Therefore, we perceive that stoplights are red more often than they actually are.

This also happens in the economic domain, as has been extensively documented by experimental economists in a variety of "public goods" games. The basic structure of these games is as follows. You have a group of people, say 10, and you give each of them some money, say $10. Each person can then contribute a fraction of their $10 to the "pot." The money in the pot is the multiplied by some factor, say 5, and then distributed equally among the ten players. So, if no one contributes anything, everyone gets to keep the $10 they were given at the beginning. If everyone contributes the full $10, the pot has $100, which is multiplied by 5 to give $500, which is distributed back to the players, and everyone walks away with $50.

The ideal thing for the group as a whole is for everyone to contribute the maximal amount. However, the ideal thing for the individual is to contribute nothing (and to hope that everyone else contributes the maximum). For example, if I contribute nothing, and everyone else contributes $10, I walk away with $55, and everyone else walks away with $45. If I contribute $10, and no one else contributes anything, I'll get $5, and everyone else will get $15. From an "economic rationality," "Nash equilibrium" perspective, the thing to do is contribute nothing. However, this is not what happens in practice.

In a wide variety of experimental setups, what people actually do is contribute about 50% of what they are initially given. So, the typical outcome in our experiment would be that everyone contributes about $5, which makes $50 in the pot, which is multiplied to $250, and everyone walks away with $30 (the $5 they kept plus $25 from the pot). Across a broad range of cultures, ages, quantities of money, etc., people come into these experiments with a somewhat liberal perspective, as they seem to both trust the good will of the other players, and care about the results for the group as a whole.

However, if we play the game over and over again, an interesting thing happens: the average contribution gradually declines, until eventually, no one is contributing anything to the pot. Based on interviews with the participants in these games, economists believe that they understand this trend. Let's say that one person contributes $5, which is the average among the group, but some people contribute $4, and some $6. This person will not really think about the people who gave $5 of $6, but will think a lot about the people who gave $4, get pissed off, and reduce their contribution in the next round. While it is mathematically trivial that people, on average, contribute the average amount to the pot, it seems to be psychologically true that people perceive themselves on average as having made an above-average contribution.

What I think is that something analogous happens over the course of the lifetime of an individual. We meet some people who are hard working, and some who are lazy, but there is this perceptual bias that means that the lazy, selfish people we meet weigh more heavily in our developing opinions about "what people are like."

The other interesting finding from these experiments is that it is remarkably easy to reset the spiral of cynicism. If you take the participants out of the room, give them a cup of coffee, and let them use the bathroom, when they go back in, they often go right back to contributing 50% on average. So, note to Democratic lawmakers, if you can figure out how to let the country drink a collective cup of coffee and use the collective bathroom, you may find a dramatic increase in support for social programs and a progressive tax structure.

No comments:

Post a Comment