Ideas to Action:

Independent research for global prosperity


Views from the Center


As we ramp up to 2015 and (maybe) a new set of Millennium Development Goals, people are proposing things to include: universal energy access, freedom from violence, a learning goal, stronger language on sustainable development... and happiness.  The Japanese government is pushing for the inclusion of a ‘happiness/well-being measure’ as part of the Rio Sustainable Development Goals, building on Bhutan’s push to include happiness as ‘the 9th MDG’.  There will be a meeting in New York later this year to discuss the idea, bringing together some of the leaders in the field.

Happiness is a good thing.  Depending who you ask, it is the ultimate in good things.  But does that make it a good fit as a Millennium Development Goal?  The existing development goals (largely) share a number of important features: they are universally agreed as an important outcome for quality of life (think education, child health).  They are, or can be, be well measured.  And they amenable to policy change.   There’s little point in getting 182 world leaders to sign up to a goal if there is nothing they can do to help meet it.  I’m not sure happiness qualifies on all of these counts.

Is happiness widely agreed an important outcome?  Surely, yes.  It is hard to imagine many politicians running on the platform: “I want to make you less happy!”  But that does raise the second question: can it be well measured? To which the answer is: that depends how you define happiness.  Is it the answer to a poll question “on a ten point scale, where zero is miserable and ten is ecstatic, how do you rate your happiness?” or is it the often-conflated answer  “on a ten point scale, how satisfied are you with life?”  Or is happiness about a life well lived --health, friends, successful career and so on?  As it turns out, empirically, there’s a relationship between  a bunch of objective measures of a life well lived and polls of both life satisfaction and happiness.  And thanks not least to work by Carol Graham at Brookings, we know those relationships are pretty stable around the world.  But, nonetheless, they are not the same thing.   And it is hard to measure something well if there’s not even widespread agreement on what we are measuring in the first place.

Let’s assume, for the moment, that what we are proposing to measure as part of an MDG is poll answers to happiness or life satisfaction questions.  Then we get to the third question: is it amenable to policy change?  To which the answer is: not much.  Or, at least, not by much we’d really want governments to do.  Most differences in life satisfaction poll answers are due to inherited characteristics, while less than 3 percent can be explained by socioeconomic status, education, income, marital status, and religious commitment combined.  As I suggest in this CGD Essay, for a society to maximize average happiness poll answers, its most effective course would probably be to put everyone on an antidepressant-ecstasy cocktail and (given the strong genetic component of happiness poll answers) add in chemical sterilization for the naturally unhappy.  Is that really what we want out of a new round of Millennium Development Goals?

So I’m with Jefferson on this one: what the UN ought to be focused on is ensuring everyone enjoys their unalienable right to pursue happiness.    Making happiness an end in itself is not such a good idea.  That means emphasizing the importance of happiness in a renewed Millennium Declaration would be great.  But suggesting we can or should raise average global happiness on a ten point scale X percent by 2030: not so much.


CGD blog posts reflect the views of the authors, drawing on prior research and experience in their areas of expertise. CGD is a nonpartisan, independent organization and does not take institutional positions.