Should We be Happy with Research on Happiness?

What does research on happiness tell us?:

Don’t ask the state for happiness, by Helen Johns and Paul Ormerod, Commentary, Financial Times: The idea that government policy should be focused more explicitly on promoting happiness has been gaining support. Proponents of this view argue that happiness indicators, based on surveys that purport to measure how happy people feel, have stagnated over decades. An important reason is that governments have aimed to maximise ... gross national product, rather than a more holistic indicator of welfare based on happiness.

This premise is clearly false. Politicians have always sought to achieve many things that are not designed to increase GNP. The most recent public service agreements on the British Treasury website, for example, spell out government commitments to ... increase participation in the arts...

A decades-long flat happiness trend could be showing that government policies in general fail... But this would be a depressing conclusion. Instead, happiness advocates make a scapegoat out of GNP and argue that economic growth is irrelevant or detrimental to happiness.

The alternative view is that the happiness data over time contain little or no genuine information. ... Indeed, they show no correlation with a whole range of factors that might reasonably be thought to improve well-being, such as a massive increase in leisure time, a tendency to live longer and a decline in gender inequality.

Income inequality is often claimed to be a strong determinant of happiness, and this “fact” used to argue for more progressive taxation. Yet we do not see any change in recorded happiness when inequality goes up or down. ...

Government attempts to increase measured happiness, rather than making life better for us, may well do the opposite: create arbitrary objectives that divert civil service energies from core responsibilities; give many people the message that happiness emanates from national policy rather than our own efforts; and create pressure for government to appear to increase an indicator that has never before shifted systematically in response to any policy or socioeconomic change. These are exactly the mistakes of the target-driven mentality that now pervades the British public sector. ...

More sinisterly, the happiness view of the world has tendencies that are inherently anti-democratic. The expert with his or her clipboard and regressions knows better than ordinary people themselves what makes them happy. So local democratic or individual decisions can be overridden with a clean conscience. ...

Government does not fail because it does not measure happiness; it fails when its energies are misdirected on the basis of poor quality information.

Should these data be used to draw conclusions such as government intervention may not improve well-being and may actually make things worse? John Quiggin at Crooked Timber has written about the data used to make these assessments:

What’s wrong with happiness measurement?, by by John Quiggin, Crooked Timber: ...For those who came in late, and probably didn’t imagine economists ever thought about happiness, the crucial finding is that “Cross country data shows pretty consistently that on average happiness increases with income, but at a certain point diminishing returns set in. In the developed world, people are not on average happier than they were in the 1960s.”

The data that supports this consists of surveys that ask people to rate their happiness on a scale, typically from 1 to 10. Within any given society, happiness tends to rise with all the obvious variables: income, health, family relationships and so on. But between societies, or in Western societies like Australia over time, there’s not much difference even though both income and health (life expectancy, for example) have improved pretty steadily for a long time.

I’ve long argued that these questions can’t really tell us anything, and an example given by Don Arthur gives me the chance to put it better than I’ve done before, I hope.

Suppose you wanted to establish whether children’s height increased with age, but you couldn’t measure height directly.

One way to respond to this problem would be to interview groups of children in different classes at school, and asked them the question Don suggests “On a scale of 1 to 10, how tall are you?”. My guess is that the data would look pretty much like reported data on the relationship between happiness and income.

That is, within the groups, you’d find that kids who were old relative to their classmates tended to be report higher numbers than those who were young relative to their classmates (for the obvious reason that, on average, the older ones would in fact be taller than their classmates).

But, for all groups, I suspect you’d find that the median response was something like 7. Even though average age is higher for higher classes, average reported height would not change (or not change much).

So you’d reach the conclusion that height was a subjective construct depending on relative, rather than absolute, age. If you wanted, you could establish some sort of metaphorical link between being old relative to your classmates and being “looked up to”.

But in reality, height does increase with (absolute) age and the problem is with the scaling of the question. A question of this kind can only give relative answers.