Richard Easterlin

Richard Easterlin

One of the long-accepted findings of happiness research turns out not to be true.  It is the finding that, except at low income levels, income isn’t associated with happiness.  The conventional wisdom has been embodied in what is called the Easterlin Paradox.  As described by economist Richard Easterlin in 1974, the paradox was based on a couple findings from the survey data available at the time.  On the one hand, once a nation’s per capita income got above the point at which most people could buy basic necessities, additional gains in income made little or no difference in life satisfaction.  On the other hand, within a given nation, more wealthy individuals typically expressed more life satisfaction than poorer individuals.  The typical explanation was that a person’s relative position in the financial pecking order matters more than his or her absolute wealth.  In other words, you can be happy living in a hut and walking everywhere you go, but not if your neighbor has a frame house and motorbike.  The implication for keeping up with the Joneses is obvious.  Besides, the paradox gives at least partial support to a belief that many people profess to hold (though actual behavior doesn’t always coincide with the belief), i.e. that money doesn’t buy happiness. 

It turns out that there’s no paradox after all.  In an article published earlier this year, Betsey Stevenson and Justin Wolfers of Wharton Business School pointed out that Easterlin’s conclusions were based on survey data from small and unrepresentative samples of countries.  Stevenson and Wolfers analysed the findings of surveys that have been conducted since then and found that there is a strong relationship between a country’s per capita income and the self-reported life-satisfaction of it’s citizens.  A linear relationship holds even in the counties with the highest incomes, though it is not between satisfaction and income but satisfaction and log income.   That means that the higher the income level, the more the additional amount needed for there to be an equivalent increase in happiness.  To use an example that Stevenson and Wolfers employ in their actual paper, a $100 increase in income would be associated with twenty times as great an increase in life satisfaction for a citizen of Burundi than for an American.   A New York Times article describing Stevenson and Wolfer’s study can be found here.

Wolfers wrote a series of blog entries about the findings: here is the first of his entries.  One question he addresses is why life satisfaction has been stagnant in the United States since 1972 even though real GDP has doubled since then.  The problem may lie in the recipients of the GDP increase.  The gains in income over the past few decades went mostly to the well-to-do.  Wolfers concludes, “given that these folks get somewhat less happiness per extra dollar, then perhaps it seems reasonable not to expect much of a rise in happiness.” He does note that, though the rest of us are mucking along about as satisfied with our lives as we were in 1972, the rich are more satisfied now than then. “But woe to you who are rich,” said Jesus, “for you have received your consolation.” They certainly are consoled.

There are two caveats concerning the article.  First, all the data is correlational, so it isn’t possible to conclude that it’s the increases in income that cause increases in satisfaction.  Second, the relationship with income is stronger when survey takers are asked about life satisfaction than when they are asked about happiness.  I’ll comment on the differences between life satisfaction and happiness in a later post.