Arnold Kling  

Surveys and Happiness

Ricardo's Difficult Idea Elude... The Morality of the Market...

I have already given so-called "happiness research" a pretty hard bashing. But Tyler Cowen thinks that there is something to it. He links to a paper that says that people who work for nonprofits are happier than people who work in other occupations.

To me, this shows the foolishness of the survey exercise. I hate working for a boss, and bosses at nonprofits tend to be less rational and less competent than bosses at commercial enterprises. So, even if a survey says that on average people are happier working for a nonprofit, there is no basis for concluding that I would be happier working for a nonprofit.

Bernie Saffran, my undergraduate economics professor at Swarthmore College, used to quote George Bernard Shaw: "Do not do unto others as they should do unto you. Their tastes may not be the same."

Do not use "happiness research" to tell me what to do. My tastes may not be the same.

For Discussion. Economists say that it is not possible to make interpersonal comparisons of utility. To what extent does "happiness research" depend on the ability to make such comparisons?

Comments and Sharing

COMMENTS (4 to date)
Peter Gallagher writes:


Isn't "happiness research" just a sample of opinion. I doubt that you need to make any assumptions about comparability of personal utility to record and analyse reports of happiness. Nor do you need to make any assumptions about the truthfulness of individual reports in the sample or about the capacity of the individuals to inspect their own state of happiness.

Your self-location way out in the tail of the (normally distributed?) sample on workplaces is not a total surprise.

But how did you feel about the regression (?) showing increasing job satisfaction with age? I found it both credible and (sort of) comforting.

Best wishes,


Gautam writes:


I completly agree with the technical reasons why economists cannot indulge themeslves in interpersonal utility analysis without significant methodological issues with the deifinition of Utility itself.

But people indulge in inter-personal comparison all the time, and there are few people who stick to their own unadulterated Utility functions, many people interpolate socially acceptable behaviour into what they really want, maybe that is the cause of "Mid-Life Crises" and the new "Quarter Life Crises", a conflict between Individual Utility Schedules and some subjectively observed Social Utility Schedule.

Then again I am just speculating here, and very off handedly at that, but inter-personal utility analysis is a result of social interaction, so we can't really echew it altogether, though it maybe wrong in a purist Theoretical sense.

Please forgive an ignorant comments on my part.

John writes:

Summing and averaging happiness without some indication of intensity seems to be an exercise in futility. Having not read the paper, I don't know if there was some happiness scale used or if responses were binary but I would discount any study that does not have some pricing mechanism where choices are costly to the participants rather than just a general survey.

vs ramachandran writes:

"For the system of economic science the main importance of this theory lies in the fact that, if distribution can be described by means of the social marginal utilities of the factors of production, it is not necessary, for that purpose, to enter into a theory of prices. The theory of distribution follows, in this case, directly from the law of social value. This theory, indeed, seems to be the starting-point of the concept of social value and the main theoretical reason for its introduction; and it helps to set forth all economic phenomena, and especially those of wages and interest, in a very simple manner -- one that is much more lucid and attractive than that derived from an intricate and cumbersome theory of prices." -joseph schumpeter, on the concept of social value1

"Happiness equals consumption divided by desire." -paul samuelson, economics2

the concept of additive utility has been easily dismissed for well nigh a century, ever since schumpeter's lucid evaluation of social value and the well-intentioned (but ultimately misguided) project he sketched for it to remake "economic science" (and presumably human welfare).

although providing an important theoretical foundation for economic study, it has been resistant to attempts at quantification, almost to the point (like dark energy) that its existance has been denied. if it cannot be measured afterall, then it cannot be real, or it must not matter.

who knows what desire lurks within the heart of men? certainly that is not for economists to say, yet this fundamental question (like why are we here?) that motivates us all as economic actors is central to behavioural inquiry.

that economics has not even attempted an answer, i think has been one of its failings. moreover if it is not addressed, i believe economics will be relegated ever further into obscurity. thankfully then, interdisciplinary inroads are being made.

while still at the fringe, work by daniel gilbert3 and richard layard4 for example (on affective forecasting and using neuroscience to scientifically study happiness, respectively) are fascinating attempts at making economics relevant to everyday people again.

intriguingly, we can use MRI imaging technique5 to actually see what desire looks like within the brain, while affective forecasting is complementing behavioural economics in a long overdue trend to root "economic science" in empirical, falsifiable (experimental) research.

4. a. b. c.

Comments for this entry have been closed
Return to top