Saturday, September 3, 2011

3/09/2011: Returns to Beauty

A fascinating study by Daniel S. Hamermesh and Jason Abrevaya, titled "BEAUTY IS THE PROMISE OF HAPPINESS?" was released yesterday by the NBER (Working Paper 17327 This is, folks, economics at its best: great question, great analysis and something that is way more relevant to life than banks, property and/or Obama's administration again blasting through the debt ceiling (oh yeah, they did that this week too - here).

Here's what I am talking about: "We measure the impact of individuals’ looks on life satisfaction/happiness. Using five data sets, from the U.S., Canada, the U.K., and Germany, we construct beauty measures in different ways that allow placing lower bounds on the effects of beauty. Beauty raises happiness: A one standard-deviation change in beauty generates about 0.10 standard deviations of additional satisfaction/happiness among men, 0.12 among women. Accounting for a wide variety of covariates, particularly effects in the labor and marriage markets, including those that might be affected by differences in beauty, the impact among men is more than halved, among women slightly less than halved."

"There are two main general mechanisms through which people’s looks could affect their satisfaction/happiness. The first is through the many channels that have been shown in the beauty literature to offer routes by which beauty affects economic outcomes. These indirect effects may be at least as important as the direct effects of beauty on satisfaction/happiness — the halo that good looks might impart to a person independent of the effects of beauty on any market-related outcomes. In the main economic exercise in this study, ...we decompose any measured impact of beauty on satisfaction/happiness into its direct and indirect components and thus examine the extent to which any beauty-happiness relation works through markets."

Hence, "The main economic question in this study is whether the effect of beauty on
satisfaction/happiness works through markets:
  • How much of the effect is direct—with people who are otherwise identical in every respect being happier/more satisfied than their less good-looking peers?
  • How much is due to the fact that beauty enhances one’s outcomes in various markets, including the labor and marriage markets?"
And here are the conclusions: "Taking the results at face value suggests that roughly half of the gross effect of looks works through the marriage and labor markets". However, "the available data do not allow us to account for impacts in other markets. As just one example, there is growing evidence that beauty confers benefits on good-looking borrowers in lending markets... Moreover, our proxies for the outcomes in the labor and marriage markets that are affected by beauty are far from perfect. It thus seems fair to conclude that the expanded estimates suggest that the direct effect of beauty is at most one-half of the total effect, and perhaps much less. The majority of the impact of beauty on satisfaction/happiness appears to be economic—through its effects on outcomes in various markets."

Yep, being beautiful pays. And hence, investing in beauty is a big business.

One other interesting point - notice that men-women differential in 'returns' to beauty are 0.1:0.12 - not exactly a massive one. "The results only slightly greater gross effects of beauty among women than among men. The direct effects, however, are larger among women, with relatively less of the impact of beauty on women’s satisfaction/happiness working through markets. This gender difference may explain why, although most of the studies measuring the impact of beauty on earnings find larger effects among men than among women, laypeople believe that looks matter more to women."

I will have to add this to my reading lists for Investment Analysis and Economic Risk courses... Simply brilliant!


Anonymous said...

good stuff. Did my thesis on the quest for happiness, I can see where beauty might help. However, its beauty on the 'inside' that matters.

well done.

Isn't Constantine a genius!!!!

fearghal said...

Interesting Post.

I'd like to raise a small but important point; isn't the study conflating beauty with attractivenss?

The two are different, yet no distinction seems to be made in the paper nor in your post. We know from experience that while we can collectively agree on attractiveness, beauty is more personal.

It seems reasonable that good looks/attractiveness- based on aesthetic rules and ratios such as symmetry and proportion- might lead to happier a life or a higher statement of happiness. But does this hold true for beauty? How can you measure the intangible immaterial aspects of beauty? Beauty goes beyond the visual and material how could you work out the returns to that?

Beauty is in the eye of the beholder; an idiosyncratic thing it changes from place to place and among people.

So, is it appropriate, to say there's "returns to beauty" when really you mean there are returns to attractiveness?

Like I said, just a small but important point. Arguing lexical nuance isn't pedantry rather the semantic due dilligence required of sensitive social research. It's an interesting study, but surely economics "at its best" shouldn't purport to say any more about the world than it is qualified to? And beauty is not a topic that economics is well equipped to speak about.

TrueEconomics said...

Thanks for the comment - just a note: it's not my research (wish it was, it is that good, imo)... so the genius title goes to Daniel S. Hamermesh (and his co-author Jason Abrevaya). :-)

TrueEconomics said...

Fearghal, I agree. It is a highly constrained and 'stylized' version of 'beauty', which in itself is not measurable directly. And you are absolutely correct - your comment is neither pedantic nor trivial. Instead, it touches on a much bigger issue. Economics (econometrics at least) can't handle many concepts that represent open sets of criteria which contain often contradictory and uncountable sub-sets of characteristics, like "beauty" or "love" or "quality" or general "value", etc. This does point to inherent constraint on economic mode of thinking, compared against philosophical or artistic exploration (or for that matter pure mathematical).

It is, however, very admirable of some economists (e.g. Gary S. Becker and others) to attempt to tackle these concepts, to deploy economic modeling and econometrics to add different context to such concepts.

This is why I love this study. Not because it is conclusive, even within the context of metrics it assesses, but because it is daring, touching on something greater than pure 'economic' variables, relating that 'greater' to something we, economists, traditionally deal with (e.g. labor markets etc).

Again, the point is not to measure - to identify - the concept of 'beauty' but to relate it sub-components that are somewhat measurable to specific outcomes that are traceable.

Great comment, by the way.

Joe said...

Constantine, I wish you could crunch the numbers on the statistical difference between boys and girls in Leaving Cert results. and the press say girls do better but there are more girls in the population (slightly) and results differ between subjects. Please do an analysis and what implications it has for society and the economy. For example, less girls do higher lever maths than boys yet a slightly higher %age of them get the honour. I just want your take on where the truth lies and where the bullshit lies.

Anonymous said...

I think it would be very difficult to handle confounding and bias in a study like this. If you are economically better off, it is likely you will have had better educational opportunities and will be more likely to engage in practices that will enhance your health and ultimately 'attractiveness'or 'beauty' (excercising, not smoking, taking an interest in your appearance etc). But presumably they adjusted for these in their multivariate model? but my guess would be they would be very difficult to capture.

Interestingly though if you look around at some of the political and business leaders we have many would not be considered traditionally 'attractive' or 'beautiful' in fact I am struggling to think of any particularly attractive ones! obviously there are other factors more predictive of economic success and my guess would be if you put them in the model the beauty variable would lose significance! I would love to see how they constructed their model and if we could do a similar one in Ireland.

Fearghal said...

@constantine agreed.

Bold research is exciting, and crossing boundaries should be encouraged. And we certaintly shouldn't shy away from asking difficult questions nor answering them with novel approaches.
I found Becker's work both intellectually intriguing and personally disturbing in equal measure as an undergrad.

In the liminal space of novelty its especially important that we provide caveats stating constraints and limitations. Economics provides a potent suite of conceptual tools, to be sure, though this intellectual fortitude strenght can be de-based by the insensitivity and hubris of mis-application of inappropriate tools and the mis-recognition of all problems as potentially economic problems. Easy and simple answers to complex messy questions- tempting as they may be- are rarely correct.

Sometimes economists, and mostly the public at large, forget this.

Thanks for replying : )