How do you really feel?

Source: The News Tribune (Tacoma, Washington)/ Getty

The World Isn't Ready for Gross National Happiness

Noah Smith is a Bloomberg View columnist. He was an assistant professor of finance at Stony Brook University, and he blogs at Noahpinion.
Read More.
a | A

Humanity has puzzled over the meaning of happiness for millennia, but now it’s being asked by economists. In recent decades, researchers have become more interested in happiness as a potential alternative to gross domestic product as a measure of national well-being. 

GDP is, in general, a reasonably good measure of living standards. Most of the people living in countries with high GDP will have cars, televisions and plenty of food, while many of those in countries with low GDP will have little or none of these. But GDP isn't a perfect measure. It leaves out the value of leisure, or time spent with one’s family and friends. It doesn’t measure our health, physical or emotional. A rancorous divorce will not subtract from GDP, and a religious experience will not add to it.

GDP 

Many people have therefore looked into the idea of measuring living standards with “gross national happiness.” Though that phrase was coined by the King of Bhutan, the idea has spread far beyond that tiny mountain nation. A number of U.S. cities and states have looked into constructing “happiness indices” for their populations. 

Meanwhile, academic economists have been ramping up happiness-related research. One prominent example is a 2013 paper by Betsey Stevenson and Justin Wolfers, which shows evidence that as countries’ income increases, their happiness keeps going up -- something that economists had previously not believed. My doctoral adviser, Miles Kimball, has also done extensive research on happiness, and has created a theory of how happiness relates to more traditional economic concepts like utility. 

The focus on happiness represents a philosophical shift for the economics field, but not necessarily an unwelcome one. Economists’ traditional measures of well-being are based on utility, or the degree to which people get what they want. If I want a burger and I get it, my utility goes up -- and, according to standard economics, I am therefore better off. It’s a fundamentally libertarian, individualistic idea, because it says that people ought to get what they want. 

But suppose the burger doesn’t make me happy? Suppose I feel bad after I eat it, because I broke my diet? Even though I might choose to eat burgers over and over, if I’m unhappy each time I succumb to temptation, am I really better off? Economists who study happiness have begun to entertain the notion that perhaps what matters isn’t the degree to which people get what they want, but how much they like what they get. Good emotions may be more important than satiation of desires. 

That’s not a crazy idea. There’s one huge problem with happiness research, however. There is really no good way to measure what people are actually feeling. 

When economists use the term “happiness,” what they really mean is “self-reported happiness.” In other words, they just ask people some question along the lines of “On a scale of 1 to 10, how happy are you?”, and mark the answer down. Sometimes they ask about retrospective happiness (how happy you’ve been over some recent time period). Sometimes they ask about “life satisfaction.” But it all really comes down to simply asking people how they feel. 

There are two big problems with that. First of all, different people respond differently to questions about their happiness. People from one culture might feel social pressure to claim that they’re happy all the time, while people from a different culture might feel ashamed of admitting their happiness to a survey researcher. These differences exist not just across cultures, but across individuals and time periods as well. 

The second problem is that people’s evaluation of their own emotions may be relative rather than absolute. Psychologists have long known about a phenomenon called “hedonic adaptation,” in which life events -- a divorce, a new job, winning the lottery -- cause only temporary increases or decreases in self-reported happiness. 

It’s perfectly possible that the emotional changes caused by these life events are long-lasting, but people simply forget how happy or unhappy they were before the event. The emotional state I call a “7” on a 1-10 scale today may be a great deal better or worse than what I would have called a “7” two years ago. But happiness surveys of the type used by economists will record me as being just as happy now as before, even though my true emotions have changed. 

If hedonic adaptation is driving most of people’s responses to surveys, then we should see little correlation between self-reported happiness and emotionally driven behavior. In fact, that is what recent Nobel-winning economist Angus Deaton and his co-author Anne Case found when they looked at the correlation between self-reported happiness and suicide:

Suicide is the ultimate act of desperate unhappiness...In a recent study...we find that some of the factors that correlate with happiness also correlate with low suicide rates, but that just as many do not...The lack of any clear relation between suicide and happiness [means that] perhaps we should...be cautious [about] giving too much weight to self-reports of life satisfaction.

If self-reported happiness can’t even predict how likely people are to kill themselves, then there are big problems with the research methodology. Emotions are incredibly important, but they are also complex, subtle and poorly understood. Until we find a better way of measuring them, I don’t think economists or governments should rush to replace traditional measures of well-being with survey measures of happiness.

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

To contact the author of this story:
Noah Smith at nsmith150@bloomberg.net

To contact the editor responsible for this story:
James Greiff at jgreiff@bloomberg.net