Thursday, December 23, 2010

The Fight Over Happiness

There’s some interesting debate concerning happiness going on out there. There have been a number of studies that show self-reported happiness hasn’t increased much when a country’s GDP grows (Easterlin’s work is key in this regard). Then we look at the chart above (reproduced from the Economist’s magazine’s daily charts and at Justin Wolfers’ work) we see a pretty clear relationship between GDP per capita and happiness. So what’s going on here? Why such conflicting results?

There are, of course, the usual arguments about econometric technique and data sources and the like. This, unfortunately, is standard with results that run counter to somebody’s religiously held opinion. But let's be a little less cynical.

There could be a prospect theory style reference point problem with the time series work. People continually reset their happiness based on the income or lifestyle they are used to. While gains are adjusted to very quickly, losses may have a much greater impact on reported happiness. Having running hot water once a week would have made somebody deliriously happy 100 years ago, but now we get grumpy if we have to cut our 20 minute soak under the shower head short. I’d like to see some of these happiness studies done right about now or a year ago in the US. I’m guessing you’ll see a fall in happiness compared to when things were going great. Having the benefits of economic growth may not make us report greater happiness over time, but not having them definitely makes us unhappy.

But there’s another possible problem with both approaches. It could be that most of the people talking about this issue have got the supposed causation backward. The hypothesis being tested is that more income (GDP per capita) makes people happier. If the causation runs this way we would expect an increase in income to increase happiness within a country over time and countries with higher incomes to have higher levels of happiness, hence no conflict about what the data is saying. But what if the causation actually runs the other way?

What if what we’re actually seeing in this data is the idea that people who are happier are more productive and can generate greater levels of output, all else being equal? This would explain why we see a relationship between levels of happiness and a relationship between growth and happiness over time without showing increased levels of happiness when an economy grows.

This is a definite possibility, but I still think economic opportunity (money) is central to happiness. If you disagree, I’d be happy to accept your money. It would make me happy just to help you out.


Lars said...

Ben Affleck's character in the movie Boiler Room said it best: "You think money doesn't buy you happiness? Here's the keys to my ferarri, look at the fucking smile on my face."

At the end of the day, money can be utilized as a way of gaining happiness than the "non-money" that poor people have. Really, someone is going to try to tell you that a starving family in rural India is fantastically happy even though they can't afford to buy a meal?

Ultimately (and yes you mentioned this) the methodology is pretty weak, as the most unhappy are very unlikely to be able to answer any survey at all (either they're currently in the process of dying or are dead).

A better measure for satisfaction in life should be mortality rate and then further broken down into suicide rate.

That can tell you a hell of a lot more about a country's satisfaction levels.

economistatlarge said...

I'm not sure about the mortality rate or suicide rate as an indicator of happiness. I assume you mean low mortality and low suicide rate are indicators of greater happiness. One way to prolong life and reduce mortality is to reduce your calorie in take to neat starvation levels. I don't know how happy this would make most people. How happy are you when you're hungry? Me not so much.