Arnold Kling  

Surveys Vs. Revealed Preference

PRINT
Protesting Economic Education... Comments of the Week, 2003-03-...

Richard Layard uses survey research and some fancy philosophical footwork to argue against conventional wisdom in economics and in favor of a nanny state. Some quotes from the series of three lectures:


People in the West have got no happier in the last 50 years. They have become much richer, they work much less, they have longer holidays, they travel more, they live longer, and they are healthier. But they are no happier. This shocking fact should be the starting point for much of our social science.
...First, I compare what I have with what I have become used to (through a process of habituation). As I ratchet up my standards, this reduces the enjoyment I get from any given standard of living. Second, I compare what I have with what other people have (through a process of rivalry). If others get better off, I need more in order to feel as good as before. So, we have two mechanisms which help to explain why all our efforts to become richer are so largely self-defeating in terms of the overall happiness of society.
...Let me give you two examples where in both cases people became objectively better off but felt subjectively worse. One is the case of East Germany where the living standards of those in work have soared since 1990, but their level of happiness has plummeted because they now compare themselves with the West Germans, rather than with other countries in the Soviet bloc. Another case is women, whose pay and opportunities have improved considerably relative to men, but their level of happiness has not. Indeed in the US women’s happiness has fallen relative to men’s, perhaps because they compare themselves more specifically with men than they used to, and are therefore more aware of the gaps that still exist.
...Rivalry and habituation lead to a quite different conclusion. They tell us that in an efficient economy, there will be substantial levels of corrective taxation.

Layard treats people's responses to survey questions such as "would you rather earn $50,000 in a world where others earn half that or earn $100,000 in a world where others earn double that?" as measuring objective reality, while taking people's actual choices as socially constructed, due to artificial influences, such as advertising.

I have a hard time letting go of my biases as an economist. If people really wanted to live among others who earn half of their incomes, Americans would be flocking to Mexico rather than the other way around.

For Discussion. Layard reports survey-based research that shows that women devote less than 15 minutes a day to sex and yet report that sex is what makes them happiest. Which of those two facts is more likely to be "socially constructed" by the way that sex is treated in the media?


Comments and Sharing





COMMENTS (3 to date)
JT writes:

I have a snide comment about Layard's thinking. If extensive government intervention in personal income is so desirable, why is practically everyone I know who works in gov't & non-profit (work environments with more equality) so much less happy than people who work in the for-profit world? Just an empirical observation.

I don't like how he is not controlling for other causes behind the decline in women's happiness.

Jim Glass writes:

This was a Robbins Lecture? They used to be good stuff, what happened?

"People in the West have got no happier in the last 50 years. They have become much richer, they work much less, they have longer holidays, they travel more, they live longer, and they are healthier. But they are no happier."

Hmm, average world life expectancy is up 50% over 50 years, so I'd imagine the average life has 50% more happiness in it. But I guess that's not how he's counting. How *is* he counting?

Even in the advanced countries, it's up about 15%. And the rates of debilitating disease in old age are *way* down -- senior years are much healthier. He's not "happy" about that for himself, if you ask him? I bet he is. ;-)

"This shocking fact should be the starting point for much of our social science."

Well, of psychology maybe, but surely not economics.

And what's "shocking" about it? It's *old* psychology 101. Literally, that was the only psych course I took in college mumble years ago, and I still learned from it that happiness doesn't correlate with income or wealth.

More recently, there's a lot of research that says people have their own individual "set points" for happiness that they always return to in the long run, come what may, good or ill. E.g.:, my first Google hit on the subject...

~quote~

Happiness or subjective well-being (SWB) ... varies from time to time about an average value that is characteristic of the individual.

My colleague, Auke Tellegen, and I have measured SWB on hundreds of pairs of middle-aged twins .... [and] obtained estimates of average SWB by repeating these measurements after intervals ranging from 4 to 10 years ... the heritability of the set-point or mean happiness level is about 80%...

Psychologists David Myers, at Hope College, and Ed and Carol Diener at the University of Illinois have been studying SWB for years with remarkable results. They find first that SWB is essentially unrelated to socioeconomic status, to income, to level of education, to gender or to race...

The Illinois researchers have shown that the effects on current SWB of both positive and negative life events are largely gone after just 3 months and undetectable after 6 months.

A happiness reading on a victim of a spinal injury -- or on a winner in the lottery -- taken a year after the event, is likely to give about the same value obtained before the event...

http://www.psych.umn.edu/psylabs/happness/hapindex.htm
~end quote~

So it is really true, money doesn't buy happiness by the very nature of things. Top economists don't know this???

Which gets us to a basic point: Economics is about *welfare*, not happiness. Rising income correlates *very* strongly with reduced child mortality, longer life, and healthier life, even in the richest societies.

Isn't that enough? Why should it be about "happiness"? Only an ivory tower academic could forget totally the *welfare* benefit of income -- which is literally life and death, even in rich countries -- and say income policies should be directed by "happiness" objectives.

Beyond that, his survey logic is abysmal. E.g.: He says East Germans report being less happy now than they were under communism, in spite of being richer now, because they compare themselves to even richer West Germans. Fine. Now ask East Germans if they would be more or less happy than they are now if they had to go back to their former lives under communism. I know enough former East Germans to tell you their answer -- "less! less! and thank God such nightmares cannot come true!" Conclusion: If the East Germans had to go back to their former lives they would be less happy than in their former lives. And if they returned to their former state when they were happier than they are now they would be less happy than they are now.

The trick is "happiness" has no objective meaning but many subjective ones, and he's conflating them all for his purposes.

Is *he* "happier" now in his day-to-day affairs than he would be if his life expectancy was 15% shorter? No. Is he "happy" that his life expectancy has gone up 15%? You bet!

"Happiness" is something to be defined and studied by psychologists. It's not relevant to economics and it is dangerous to bring it into economics. People have tried in the past, back to the utilitarians, with "not happy" results.

"Another case is women, whose pay and opportunities have improved considerably relative to men, but their level of happiness has not. Indeed in the US women’s happiness has fallen relative to men’s, perhaps because they compare themselves more specifically with men than they used to, and are therefore more aware of the gaps that still exist."

He's building a lot on that "perhaps". Perhaps women's happiness has fallen relative to men's (if it has) not because they are now "more aware" of the smaller wage gap that still exists -- note how patronizing this is, women would have had to have been pretty stupid to be "less aware" of a much larger wage gap in the old days -- but because, e.g., increasing income is associated with greater job responsibilities that cut into family life.

Yet obviously women prefer the higher income to extra family time because nothing's easier than giving up a job for more family time, even if the higher income they prefer to happiness makes them less happy. Though you should be happy to take something you prefer to happiness. Ouch ... Leave it to the psychologists.

"I ratchet up my standards, this reduces the enjoyment I get from any given standard of living"

So if it really is the ratcheting up of his standards that reduces his enjoyment, he should reduce his standards. Simple.

".. in an efficient economy, there will be substantial levels of corrective taxation."

Efficient as to what: welfare or happiness? How does he propose to increase "efficiency" of happiness? By taxing everybody to ratchet down their living standards? He can't even *define* happiness.

Moreover, "corrective taxation" will never be near enough to deliver happiness. That will take more much general legislation enacted by the enlightened political powers. And we've had some very bad experiences in the last 100 years with regimes that claimed to legislating happiness for all, peoples' utopias. This is not only a stupid idea but a dangerous one to bring into political economics.

A Robbins lecture is favoring the legislation of happiness as policy! The rude natives of sci.econ on usenet have a Dumbassy award they give out periodically to particularly deserving postings. If this was there, it could be the first Robbins lecture to win one!

Sorry to run on at such length but this one really teed me off.

kevin quinn writes:

Jim Glass is write to champion objective rather than subjective measures of well-being. But he is dead wrong in thinking that that is what economics is about! The First and Second Welfare Theorems claim, in effect, that free markets lead to outcomes that cannot be improved on by making somebody or bodies happier without making someone else less happy (happiness is preference satisfaction, period). Layard is right to say - as Robert Frank has been saying for a much longer time - the pursuit of status etc destroy these welfare theorems. I am working right now on a paper that argues that Smith - despite the perception that he fathered the First Welfare Theorem - a) had an objective theory of well-being and b)knew all about the phenomena Frank and Layard talk about, but praised emulation, the pursuit of status,etc. because, despite the negative impact they had on happiness, they raised well-being in an objective sense. So, back to the future-

Kevin Quinn

Comments for this entry have been closed
Return to top