"I'd rather be a really good one-term president than a mediocre two-term president."
A rare case of a politician with a backbone. Given that, I can't say I was surprised by the election outcome. No, what depressed me was the lacking substance of arguments. The American nation strikes me as similar to a group of overweight people who at their first weight watchers meeting chants "Yes, we can" and cheer upon change. But when change is staring back from the dinner plate, and change on the scale leaves waiting, they realize change doesn't come easy. And the vast majority of them still doesn't know the difference between social democracy and socialism. Clearly, the world would be a better place if everybody would read my blog ;-)
Anyway, to some extend I don't care very much how the Americans organize their society. I think they're not fully using their potential, and find that a shame, but after all it's their decision what they put on their plates and shovel down their throats while I, well, I live in Sweden. And that brings me to one of the most amusing studies I've come across lately:
- Building a Better America – One Wealth Quintile at a Time
By Michael I. Norton and Dan Ariely, PDF here
Michael Norton, from Harvard Business School, and his colleague Dan Ariely, from Duke University, asked a random sample of US citizens what wealth distribution they think is ideal. In 2005, they surveyed 5,522 people. Asked for their voting pattern in the 2004 election, the sample reproduced well the actual voting result. The survey respondents were given a definition for wealth so there was no ambiguity. Then they were shown three pie charts. Each slice of the pie represents 20% of the population, from the poorest to the wealthiest. The size of the slice is the wealth owned by this group. One pie showed a perfectly equal distribution. The other two pies were unlabeled but showed the distribution of the USA and that of Sweden.
The result: 47% of Americans preferred the Swedish wealth distribution, followed by 43% for the equal distribution, while only 10% found ideal the actual distribution. Just focusing on the Swedish vs the US distribution, 92% of Americans prefer the Swedish one over their own.
[Source: Fig 1 of this paper]
It turns out that these preferences depend only very little on demographic factors like gender or whether they voted for Bush or Kerry in 2004. Considered how convinced Americans tend to be about their own greatness, this result seems somewhat puzzling. However, keep in mind that these pie charts were unlabeled in the questionnaire. The replies makes sense if you come to the next question. In that, survey respondents were asked first to guess the wealth distribution in the USA, and then chose what distribution they would find ideal. It turns out that most Americans severely underestimate the rich-poor gap in their own country, and in addition would prefer a distribution that is even more equal than their erroneous estimate. This is shown in the figure below.
[Source: Fig 3 of this paper]
Again, note how little both the estimate as well as the ideal depends on demographic factors.
This result fits quite well with previous studies which had shown that Americans overestimate the social mobility in their own country. They're still dreaming the American dream, despite its evident conflict with reality.
After I stopped laughing I started wondering what this result really means. The survey respondents are very clearly considering the present wealth distribution as not ideal. However, the wealth distribution is a fairly abstract observable. Would you have been able to accurately estimate it? My own estimate would have been considerably closer to the actual one than the average guess, but that's only because I happen to have seen the relevant numbers before.
Norton and Ariely had a good reason to ask these questions: The philosopher John Rawls proposed that justice should be identified by taking a position behind a "veil of ignorance." For that, you're supposed to imagine that you decide on a particular question - for example the distribution of wealth - and only after you've decided you'll be randomly assigned a position within that society you've just created. I've never been really convinced by that approach. It's much too heady, or call it utopian. As a matter of fact, people don't live behind a veil of ignorance and their own social status does influence their decisions. Also, it isn't only the ideal (size 4!) that's relevant but also the way to get there (diet). In fact, the way is typically the question that's more immediate and thus more prominent on people's mind.
If one just asks people what they think is ideal, you're probing their ideas about what they believe the wealth distribution means, not necessarily what they actually want. To get to the relevant point, one would have to ask for factors that actually affect their life, or are such that they have some basis to judge on. Social mobility for example, the possibilities that are open for them and their children, is a relevant factor, and it is of course related to the wealth distribution. Or, instead of asking for the distribution of wealth, maybe better ask if they think somebody's work is really worth a 1000 times more than somebody else's. Another factor, and the one that bothers me most, is that wealth means power and it means influence. How much influence on your life do you want a small group of people to have? And at which point does this run into conflict with democratic decision making?
Bottomline: This is an interesting study. It explains a lot of things about the US American attitude towards their country's income distribution and the sometimes puzzling disconnect between their wish for change on the one hand and on the other hand their unwillingness to really take the necessary steps: they believe the steps are smaller than they in fact are. However, it's not a result that should have any relevance for policy decisions because the question asked is impractical. One doesn't chose a wealth distribution first and then gets randomly assigned a place in that society. It's not how things work in real life, and it's just replacing one dream with another one. There's always the risk the dream might later turn out to be a nightmare.
Aside: Dan Ariely, one of the authors of the study, writes a blog. He commented on his own paper here.