What is the relationship between material wealth and overall well-being? That question is, after more than half a century of being largely excluded from the discourse of academic economics, coming back into fashion. Tibor Scitovsky’s The Joyless Economy led the way, but it was so far in advance of its times as to be largely ignored despite the distinction of its author. The same could be said of Thomas C. Schelling’s essay “The Mind as a Consuming Organ.” Robert Frank’s Luxury Fever drew more attention, but it had flaws in exposition that allowed even someone as generally open-minded as Jack Hirshleifer to dismiss it as “a demonstration of what happens when a good economist gets mugged by Thorstein Veblen.”
Now comes Richard Layard, in this year’s Lionel Robbins lectures. [Thanks to Brad DeLong for the pointer.] There is irony here: It was Lionel Robbins, in The Nature and Significance of Economic Science, who persuaded the discipline seventy years ago to abandon the idea of “utility” as a measurable (at least notionally measurable) quantity, comparable across individuals, and to substitute preference satisfaction as its guiding principle.
The following propositions seem to be reasonably well established:
1. Survey questions about “happiness” or “life satisfaction” seem to elicit consistent and meaningful responses, which seem to have predictive power over various objectively measurable phenomena such as morbidity and mortality, holding constant the obvious potential confounders.
2. Transient self-report measures of “happiness” and “unhappiness” seem to correlate closely with measures of localized brain activity in two areas in the prefrontal cortex, with (in normally-wired right-handed people) happiness marked by activity on the left side of the brain and unhappiness by activity on the right side.
3. Most of the variation in overall happiness appears to stem from individual-level constitutional factors rather than anything observable in the environment. Twin studies show fairly high correlations among identical twins raised apart, vanishingly small correlations among fraternal twins raised apart.
4. Being healthy, married and living with one’s spouse, and securely employed all correlate strongly with self-reported happiness. The effects of employment and employment security remain even after controlling for income.
5. Whether you’re rich or poor, it’s nice to have money. In every society, those with higher incomes or greater wealth report greater average happiness.
6. The contribution of money is relatively modest compared to other situational factors: a divorce, for example, is two and a half times as bad as a drop in income equal to a third of the average income in the surrounding society.
7. Surprisingly, the correlation between wealth and well-being observable in any society at any time, with richer on average meaning happier, doesn’t carry over across societies or over time. Over the past half-century, GDP per capita has more than doubled in the US and sextupled in Japan, without moving the average happiness score in either country. People living in very poor countries tend to be less happy than people living in richer countries, but above about $16,000 per capita — roughly half the current US level, putting Ireland and New Zealand just above the cutoff and Spain and East Germany just below it — there seems to be no consistent trend. Both Ireland and New Zealand, for example, are slightly ahead of the US in average reported happiness.
8. The explanation seems to be threefold. First, people habituate to consumption patterns; as Scitovsky puts it, pleasures (which cause active happiness by their presence) degenerate into comforts (which aren’t noticed if present but cause discomfort if absent). Anyone who spends time thinking about drug abuse will recognized the pattern. Second, people evaluate their own consumption bundles relative to standards that are sensitive to average consumption patterns; when asked “How large an income would someone in this area need to live decently?” people give answers that, over time, track closely the actual average income. Third, having more than one’s neighbors confers status, while having less is a status insult, so some of the benefit of income at the individual level is winnings in a zero-sum game.
9. Not all forms of material comfort are equally subject to habituation, norm-referencing, or rivalry. Leisure, for example, seems to be absolutely, rather than relatively, valued, at least in paper-and-pencil exercises.
Layard and Frank draw similar sets of policy prescriptions from these observations: Tax private consumption to finance public services (on the theory that having a larger car is a zero-sum game, while breathing cleaner air isn’t). Try to revise the terms of the leisure/income tradeoff to encourage more leisure. Redistribute income. Target happiness directly rather than GDP growth in making economic policy choices. Change parenting and educational practices to (1) make people more conscious of what does and does not make them happy and (2) encourage values that produce less zero-sum consumption behavior.
All of these seem to make sense, though none is bullet-proof. In particular, the logic of redistribution is that more of the consumption of the poor is non-zero-sum than is the case for the rich, which may or may not be true. (That might be a question subject to empirical study.)
Layard’s lectures, more than Frank’s book, consider explicitly some of the philosophical questions raised by this, but I can’t report that Layard is especially successful in doing so. In particular, Layard elides the distinction between transient states of feeling well (what Bentham called “pleasure, and the absence of pain”) and overall happiness or satisfaction with one’s life.
But Kahnemann has shown that, in evaluating a pleasant or unpleasant experience, people’s overall assessments, when compared to their moment-by-moment assessments, don’t even obey the principle of dominance. That is, an experience can be evaluated as less unpleasant overall even though it was evaluated as more unpleasant at each moment. It isn’t immediately obvious which evaluation ought to control. The problems if one extends this to a lifetime rather than a colonoscopy are presumably even greater.
Moreover, even life-satisfaction isn’t obviously the same thing as overall well-being or well-faring, unless, as J.S. Mill said, you’re really prepared to say that a satisfied pig is happier than a dissatisfied Socrates. Imagine that it turned out that the brain activity that correlates with moment-to-moment happiness were actually causally related to it, and that we learned how to stimulated brain activity in the key left-brain locations and suppress it in the key right-brain locations. (This is an old science-fiction speculation.) Would a life lived passively lying back and having one’s pleasure center stimulated electrically truly be a choice-worthy human life?
Layard dances past this question in his discussion of religion. It turns out that people who report belief in God report themselves as happier, even controlling for frequency of attendance at religious services. Layard, who judging from these lectures is some sort of agnostic humanist, would like to have it that “God” merely means being in touch with some principle of right action, and says that it would be useful to encourage people to establish that contact, quickly adding that no one should believe anything contrary to “his reason.” But that avoids the hard question. What if there were no God, but believing in God made people happier? Or, to get out of the metaphysical realm, what if the theory of evolution were substantially correct, but believing in a divine origin for human beings made people happier? Or if people who thought that the Earth was the center of the Universe were happier than those who had heard of Copernicus and Einstein? Or simply that (as seems to be the case) people who optimistically misjudge their own actual capacities and the opinions of other people about them are happier than those with more accurate self-knowledge? On what grounds would Layard say that such beliefs were a bad thing?
There’s a simple answer to that question, I think. Pleasure is not the sole good. Dignity, for example, is also a good, and it’s undignified to believe things that are palpably false just because it makes you feel good. (*) [See “clarification” below.]
What’s missing from Layard’s analysis is any sense of the possible distinction between “hedonics” and “eudaimonics.” Merely going back to Bentham cannot really be the way forward.
Still, from a larger perspective, all this is quibbling. To shake economics loose from its fixation on the satisfaction of subjective preferences, and refocus its attention on the question on the material causes of human well-being, is a great service, and Layard and Frank have begun to lay the foundation of a more useful economic science.
Clarification I didn’t mean to imply that belief in God was false; that question would get me in to metaphysical waters over my head. I certainly didn’t mean to say that individual believers were lacking in dignity. I did mean to say that, IF a given religious (or other) belief were false, it would be undignified to hold it, just as it is undignified to believe that the Earth is flat. Someone could have considerable personal dignity and still hold such beliefs, of course, but I claim that the thing itself — believing what is not the case, and in particular embracing beliefs that make one comfortable rather than those supported by reason and evidence — is undignified, like speaking one’s native language poorly, walking around in pubilc with one’s fly open, or farting in church.