Utilitarianism, Happiness, and Optimism

I’ll admit I don’t know a whole lot about happiness research. But as a utilitarian, and as an economist trained in the neoclassical paradigm, I’m rather skeptical. I’m skeptical, in particular, of attempts to equate quantitative, empirical measures of happiness with utility in the philosophical sense or with “true” happiness in any quantitative sense.

A case in point is the frequent finding that optimism is conducive to happiness. Suppose that a study of long-term investors found that optimistic investors obtained better returns. That wouldn’t surprise me at all, but I wouldn’t interpret it to mean that optimism is conducive to better investment results. Indeed, I would conclude exactly the opposite. People are naturally risk-averse; consequently more risky investments must have higher expected returns in order to attract investors. If an investor has average preferences but is “optimistic” in the sense of overestimating expected returns, that investor will choose riskier investments, because he or she will judge (wrongly) that those investments are worth the risk. Similarly, people who have optimistically biased expectations about their lives will be willing to take risks that, on average, result in better outcomes. This doesn’t mean that optimism is good; it just means that the measure of outcomes is not accounting sufficiently for the severity of the bad outcomes among the minority of optimists whose risks do not pay off.

I don’t doubt that optimism has its virtues. Surely certain optimistic biases can be good for the individual if they compensate for other pessimistic biases. And the optimistic bias of entrepreneurs, even if it isn’t good for them, is good for the rest of us, because it compensates for the market’s inability to offer them full diversification of risks. I won’t rule out the possibility that optimistic bias directly affects happiness to an extent that might outweigh the effect of the irrational behavior it produces. But any study that purports to demonstrate that point empirically would require a subtlety of design not easy to achieve.