Friday, January 18, 2008

Relative vs Absolute Wealth and Revealed Preference

In class today the question of concern with relative wealth came up and one student mentioned an article according to which a majority of people said that given the choice between two alternatives, in one of which their absolute wealth was higher but other people's was higher still, they would prefer the alternative with lower absolute but higher relative wealth. I don't have the cite so can't give the actual numbers for the alternatives.

Economists prefer to measure preference by actions rather than words. It occurred to me that we have such evidence for this question. Lots of people choose to migrate from one society to another. If you go from a rich society to a poor society you are likely to substantially increase your relative position, since you will bring with you both wealth and human capital that are high compared to the average in your new home. If you go from a poor society to a rich society, the opposite can be expected.

I don't have data, but my impression is that migration from a poorer to a richer society is much more common than migration the other way. If so, that provides evidence, at least for the alternatives that migrants face, that absolute wealth is more important than relative wealth.


TGGP said...

I've made similar points about the "foot vote". I think Alex Tabarrok makes a relevant point here.

Anonymous said...

If we actually favor absolute wealth but tell interviewers (or ourselves) that we prefer relative wealth, that leads to another curious conundrum... what explains the discrepancy between our imagined and our revealed preferences?

Tim of Angle said...

This article may be relevant.,0,2616027.story

Mark said...

It is a great negotiation advantage to be able to convince yourself, and anyone you're negotiating with, that relative distribution is something you value highly.

Even better if you can maintain cognitive dissonance enough to make other decisions on the basis of absolute wealth maximization.

Anonymous said...

Remember that many of the people migrating from poor places to rich places do so with the intention of sending money back to their families, to improve the latter's relative wealth.

In addition, some economic migrants hear glowing stories about salaries in the developed nations, or the Big City, or wherever, and don't fully realize that (a) that salary doesn't buy as much as it would at home, and (b) they're improving their absolute wealth but not their relative wealth.

Economics would predict that people would learn those lessons and settle down to a rational level of economic migration, but long-distance economic migration carries considerable transaction costs: a migrant living on minimum wage in a rich place may not be able to save up the capital to move back to a poor place if (s)he wanted to.

Stuart Buck said...

A lot of relevant citations are discussed here.

One problem is that it's not clear that people are really rejecting a higher "absolute wealth" in these studies. There's no absolute value of money. So if you ask people whether they would rather have $50,000 in a world where others have $40,000 vs. having $100,000 in a world where others have $150,000, some people might instinctively realize that in the first world, $50,000 would actually buy more stuff than $100,000 in the second world. Right?

Anonymous said...

Greg Mankiw brought up this point which I think is pertinent as well. The conventional wisdom is that people tend to have a smaller house in a wealthy neighborhood than a larger house in a poor neighborhood.

William Newman said...

tim of angle: The _LA Times_ article is interesting, but to pick on something sloppy: "If there are behavioral analogies between humans and other primates, the underlying brain mechanism driving the choice preferences most certainly dates back to a common ancestor more than 10 million years ago."

There are many examples of independent parallel evolution in nature. E.g., it would not be safe to sketch some feature of the hunting behavior of a Tasmanian wolf, the swimming behavior of a dolphin, or the eye-tracking behavior of an octopus and conclude "most certainly dates back to a common ancestor" with a non-marsupial wolf, a fish, or a mammal.

That quibble doesn't seem to directly weaken his main point, but it does seem to be evidence that he might not have bounced his ideas off experts before publishing. (And if he had, he might have learned of other interesting stuff that people sometimes tie to the argument, e.g., the way that one might expect band size to affect the strength of the payoffs to strategic behavior and reputation tracking and such.)

William Newman said...

hudebnik: I agree that phenomenon is probably part of the mix, but when you let people move freely (so, e.g., it's not too hard to bring the family), it seems to become clear that it's thoroughly dominated by other effects. E.g., my impression is that it doesn't explain either US immigration balance back when immigration was less legally constrained, or explain movement within the British Empire and whatever it became (commonwealth? intermediate forms? beats me). Really I have to stretch to think of any highly-free-movement case it might explain.

It might match some patterns from a study I read long ago about Puerto Rican immigration to New York. But if an instinctive pursuit of relative status were were the dominant factor in Puerto Rican immigration (as opposed to language factors, or tax factors, or Spanish ex-colonial culture factors, or something) then why isn't it clear in other immigration within the US? It isn't clear to me.

To the extent that we can say infer anything simple from people's footvotes, it seems to me that David Friedman's (and/or tggp's or mankiw's or whoever's) pattern dominates. One relativity preference does seem to be revealed: Stuart Buck's ratio of cost of labor to cost of living. And a preference for a nice climate is revealed. But I don't see a revealed preference for cutting off one's nose the better to look down upon one's peers.

Unknown said...

One question I have is how the survey is affected by what relative and absolute levels are being considered. If I am considering which of two raises of my current level of wealth I would prefer, that might get a different answer from whether I would prefer a raise in my absolute wealth and a drop in my relative wealth over the reverse. In the first case, I might prefer a small raise in my absolute and a big raise in my relative. In the latter I certainly prefer a raise in my absolute over a fall in my absolute, _regardless_ of what that does to my relative (though Stuart Buck's point still applies, but even if we equalize for that somehow, my point is still worth considering)

David Friedman said...

Stuart writes:

"So if you ask people whether they would rather have $50,000 in a world where others have $40,000 vs. having $100,000 in a world where others have $150,000, some people might instinctively realize that in the first world, $50,000 would actually buy more stuff than $100,000 in the second world."

Interestingly enough, that was the argument my fourteen year old son came up with when I raised the question with him.

I'm not sure why Stuart puts it as "realize" rather than "assume." Some societies are, after all, richer than others. I don't know how the question was put, but it surely could have been put in terms of real, not merely nominal, income, and that is how I would have interpreted it.

Anonymous said...

If I'm starving, and so is everybody else around, then moving someplace where I have plenty to eat and everybody else is rich is a big win.

I've seen the issue phrased as "Would you prefer to get a $5,000 raise while your co-workers get none, or you get $10,000 and they get $15,000?" Absolute levels make a difference.

Anonymous said...

I'd think people always prefer absolute wealth, and someone honestly preferring relative worth at the expensive of absolute wealth has psychological problems.

However, I don't think your migratory patterns will prove that point, unless you control for other factors (such as age, and wealth level at time of migration).

People migrate to wealthier countries to increase their absolute wealth (higher wages, larger middle class, more/better job opportunities, etc). People migrate from wealthier countries to poorer countries ***also to increase their absolute wealth*** (lower property/housing costs, lower costs of domestic staff, lower medical costs, and lower/nil taxes).

Beastin said...

I like Stuart Buck's point. I personally would choose the $50,000 scenario for his phrasing of the question.

I'm not sure that it's impossible to reconcile people's stated preferences and the migration patterns observed by David though. Suppose that people judge the value of future income against not the relative wealth of their future peers but the relative wealth of their current peers. You'd then expect migration to overwhelmingly favor rich countries. Though, if they subsequently adopt the standards of the new country, you'd expect migrants who are poor in their new country to generally be more dissatisfied with life, and I'm not sure how true that is.

If you asked me whether I'd like to be as rich as Larry Page I'd say, “Yes.”. I wouldn't say, “Oh, but then Bill Gates would always be staring down his nose at me.” Probably I wouldn't care much about Bill either, since I'd still likely be at the financial top of whatever social group I moved in.

montestruc said...

I think you hit the nail on the head. Just like the joke you had in your textbook about the two economics professors passing the sports car dealership. {First professor "I want that car!", second professor, "Obviously you do not."