The Happiness Wars Continue

Photo: iStockphoto

There’s a growing sentiment among economists that GDP is a poor measure of a country’s well-being. (See our recent podcast on the topic; also, the research of Joseph Stiglitz.) The latest fad among European governments seeking to separate the overall health of citizens from sluggish economic data is to ask them if they’re happy. The results aren’t exactly encouraging. Less than half of British adults feel they are thriving. And France now ranks as the world’s most pessimistic country, with only 15 percent saying they expect things to get better in 2011. In a recent piece for The Lancet, NYU economics professor William Easterly consolidates the flood of new happiness research with a discussion of the 2010 book, International Differences on Well-Being, a compendium of sorts on the topic. Easterly begins by reminding us of the Easterlin paradox, which stipulates that it is our income relative to our peers rather than our absolute income that determines our happiness. (This has formed the basis of the theory that people in richer nations do not tend to be happier than people in poorer ones.) And then, citing new research, Easterly goes about knocking it down:

In my view, the paradox has not held up that well, although it survives in some truncated form. The first important finding that has emerged from the new flood of data is that the original idea of “happiness” failed to discriminate adequately between two different ideas: first, emotional feelings of “happiness”, measured on a day-to-day basis; and second, a more stable, long-run satisfaction with one’s life. The first continues to be called “happiness” (although the editors of this book prefer the slightly less catchy “affect balance”), while the second is now called “life satisfaction”. Both are valid and useful concepts, but they should not be confused. The newest surveys do a good job at measuring the two separately. The Easterlin paradox turns out to hold much more with “happiness” than with “life satisfaction”.
The piece of the Easterlin paradox that really falls apart with newer data is his evidence that there was little difference between rich and poor countries on average happiness. Especially with the life satisfaction concept and a much larger sample, the differences are now recognised as vast, as shown in the chapter here by Ed Diener, Daniel Kahneman, and colleagues. As I heard a happiness researcher once state it most colourfully, the average Togolese man would be hospitalised for depression in Denmark. The non-paradoxical piece of the paradox remains as strong as it always was—within the same country, the richer are happier and more satisfied with their lives than the poor.

Related: Justin Wolfers has written extensively on this blog about happiness.

Charly Suter

Probable we should ask people of each country if the would recommend to live there and then take this Net Promoter Score (NPS) to measure and compare...

Scott M

It seems that the Easterlin Paradox would not hold up to more temporal measurements. For example, comparing the happiness of a fifty year old in 1750 to the happiness of a fifty year old in 2011 would probably drive a huge gap in happiness, as the overall wealth of one would grant more access to the technology and health-related products of 2011.


Thanks for the interesting post! Here at eZonomics, we've just run a poll asking if people think happiness affects the wider economy - with 57% of respondents answering "yes". Our full post is here We'll run the reverse question (does the wider economy affect happiness) at a later date.

Nick Verdant

Let me try to understand this. Happiness is a measure of success in competition, while life satisfaction is something deeper? Should happiness correspond to divide between the have and have nots?

Rich EconStat

In my view, the indexes should detect from a "maslow hierarchy of needs" perspective.

For example: an Iraqi who 3 years ago lived in a place where there were 3 suicide bombers per week where there are now only 1 per month may be really happy now. But, he may be happy based on low expectatation and lack of information. If any study then compares across countries it is impossible to separate these out. Even "in country", it's probably invalid - better off people have more info that their relative situation is bad versus less well off people (living in ignorance is great!).

I have the a similar issue with Quality of Life indexes, where Scandinavian countries come out well. Well, the researchers should go and live there for 2 years and see if that's what you would call "quality of life". The index should be re-titled "places where you get a better start in life so you can move somewhere else or, if not, be cared for by a good health system paid for by the fact that your net income on your marginal $ in salary is 30% and we spend the 70% well" index.

I like the concept of "happiness researchers"!


Lisa Sansom

OECD has a new metric out now too:,3746,en_2649_201185_47912639_1_1_1_1,00.html