David Leonhardt reports in the NYT on the Stevenson and Wolfers paper I blogged last week. This gives me hope that the conventional wisdom is starting to shift with the evidence. It's worth noting that although the new Gallup World Poll has been very useful, the evidence isn't really new. Here's Ruut Veenhoven in the discussion we had on happiness almost exactly a year ago in Cato Unbound:
Time series data on happiness are much improved lately and now present a different picture.[2] Happiness appears to have risen in many nations over the last forty years. The greatest increases have been observed in non-Western nations such as Brazil, Egypt, India, and Mexico, with an average gain of about one point on a scale from 0 to 10 since the early 1960s. Happiness has also risen in the eight EU nations that have participated in the Eurobarometer survey since 1973, with a gain of about 0.3 points in 33 years. A similar trend is observed in the United States, where average happiness also rose 0.3 points since the early 1970s. However, compared to the first happiness surveys conducted in the late 1940s, American happiness seems to have hardly improved.
This is, in broad outline, the same story Stevenson and Wolfers are telling.
This past weekend I was at a conference that discussed a number of papers from the past decade or two drawing on the happiness literature. It is truly maddening how the measure least likely to be informative — the trend in average national self-reported happiness — is what gets top billing, over and over again. It has been crystal clear in the data basically forever that there is a positive correlation between average income and average national happiness. And that within countries there is a positive correlation between individual income and happiness. The evidence has always been strong that money makes a significant positive difference for happiness. Upon seeing a flat trend in average happiness over time as average income rises, you'd think the right thing to do would be to ask what is wrong with THAT measure. A ceiling effect? Scale renorming as expectations rise? But no. The measure that suggests income growth really does us no good, that must be right. So let's hold that fixed and then try to explain away the significance of the strong within-country correlation by making up ill-supported just-so stories about zero-sum status races.
Now that it is increasingly clear that there is a cross-national connection between income and happiness, that it doesn't exhaust itself at $15,000, that just about every country that has gotten richer has gotten happier, and that within these countries, richer people tend to be happier than less rich people,
I'm sure we can all look forward to a new set of ingenious theories that avoid the obvious interpretation of the data, which is that, other things equal, having more money makes life better.
[UPDATE: Here's Justin Wolfers' first blog post about the Easterlin paradox and his and Betsey's paper at the Freakonomics blog.]
Will Wilkinson 