Felicific Calculus

Here’s a thought-provoking article about a new study on the satisfaction people experience while taking part in various life activities. (Thanks to Tyler Cowen for the link.) Among other things, the study shows that people obtain the most satisfaction from having “intimate relations” (I assume that means sex), socializing, relaxing, praying/meditating, exercising, and eating. They obtain the least satisfaction from commuting, working, and doing housework. I don’t find any of this terribly surprising. But the conclusion of the article set off my alarm bells:

"It's not that life circumstances are irrelevant to well-being," notes [study co-author Norbert] Schwarz. "On the contrary, we found that people experience large variations in feelings during the course of a normal day. This variation highlights the importance of optimizing the allocation of time across situations and activities. If you want to improve your well-being, make sure that you allocate your time wisely."

Unfortunately, that's not easy. When the researchers examined the amount of time spent on various activities, they found that people spent the bulk of their waking time – 11.5 hours – engaged in the activities they enjoyed the least: work, housework and commuting.

The implication is that people are failing to optimize their time, since they spend relatively large amounts of time on those activities from which they obtain the least satisfaction. But for optimization purposes, the issue is not the instantaneous satisfaction obtained from an activity, but the net satisfaction resulting from it over time. Nobody I know commutes because they find it inherently enjoyable; they commute because the corresponding gains – better living conditions available farther away – make the commute worth it. And while some people do enjoy their jobs, mostly people work (or work as much as they do) because the wages and benefits allow them greater satisfaction during their off hours. Optimizing time usage implies equalizing the net marginal returns across all activities and over time, not shifting time toward the activities that generate the greatest immediate satisfaction.

This oversight highlights a pitfall of the new “happiness measurement” literature, which attempts to measure people’s satisfaction, and the rationality of their choices, by eliciting subjective assessments of happiness from test subjects. The problem is that such assessments will tend to reflect the agents’ feelings at the moment the question is asked, rather than reflecting overall effects of activities across time. In addition, the survey responses do not necessarily measure the “all-inclusive” form of satisfaction that is relevant for judging the rationality of individual actions; instead, they may indicate judgments about more narrow forms of satisfaction, such as instantaneous hedonic pleasure. That kind of satisfaction is assuredly relevant to making good choices, but it is very far from the whole story.