Why it's hard to measure happiness
- 16 November 2010
- From the section Magazine
The government wants to measure our happiness. It won't be easy, explains Michael Blastland in his regular Go Figure column.
Did you wake up fretting over the gross domestic product? Thought not. But you might, as you gaze in the bathroom mirror, ponder your happiness or well-being.
Not simply GDP then, but GWB (general well-being), as David Cameron put it. Though Labour was taken by the idea too, like President Sarkozy of France, who recently asked two Nobel Prize winners to look into it.
Knowing whether we're making more than material progress makes sense. The problem is how?
GDP, almost all agree, is a deficient way of adding up what matters most: "Economic performance is... a means to an end. That end is not the consumption of beefburgers, nor the accumulation of television sets, nor the vanquishing of some high level of interest rates, but rather the enrichment of mankind's feeling of well-being. Economic things matter only in so far as they make people happier."
So said the UK's happiness-and-economics guru, Andrew Oswald, a smart - and cheerful - professor of economics from Warwick University.
Here's the easy way to do it. First, take a large, random sample of people and ask them a question like: "When all's said and done, at the end of the day Brian, taking the rough with the smooth and all that, rate your well-being on a scale of one to 10."
And we could stop there and add up the national total each year to see if well-being has improved.
But that would only invite more questions, like "what causes well-being to change?" We'd then have to ask lots of other questions about age, gender, marital status, employment (or not), qualifications, housing, recreation and so on.
So, marrieds score higher than singles? QED, marriage is good for wellbeing. Or maybe not. For it soon gets tricky.
Maybe happy people are more likely to marry and we had cause and effect the wrong way round. That's a risk, say researchers, which they address by tracking people over time, to see how happy they were before marrying and what difference marriage made.
But what if people are happier in their 30s than their 20s? If so, following them over the years also picks up the happiness effect of ageing - which we might falsely attribute to marriage.
So you have to find out, are 30-somethings happier than 20-somethings and, if so, can we strip that effect out of the marriage effect that may have happened at the same time? Still with me? And so on, for everything you can think of that might mess up your measurement of the effect of one thing on well-being, rather than another.
So even the easy way is not always easy. For an interesting attempt to measure cause and effect try Mappiness, a project run by the London School of Economics, which offers a phone app that prompts you to record your mood and situation.
The Mappiness website says: "We're particularly interested in how people's happiness is affected by their local environment - air pollution, noise, green spaces, and so on - which the data from Mappiness will be absolutely great for investigating."
Will it work? With enough people, it might. But there are other problems. We've been using happiness and well-being interchangeably. Is that ok? The difference comes out in a sentiment like: "We were happiest during the war." But was our well-being also greatest then?
And what's the policy response? That what we need is a war? And then you remember: we've got one - and it's not making us happy. Maybe that's because to be really happy it has to be a war of national survival. Maybe what we need to be really happy, in other words, is for our circumstances to be really miserable.
But maybe this war is improving our well-being by protecting us from terror? So values and political judgements about what constitutes well-being soon muddy the waters, and these may be different from what makes us happy.
What's more, your well-being may not be improved by the same things as mine. The fact that you like to get stoned isn't going to see cannabis in Tesco.
Our leaders have to weigh up the pros and antis on each side of any factor in 60m people's well-being. Maybe my environmental well-being - a category often discussed - is served by renewable energy, yours by hillsides free from wind-turbines.
It all begins to sound less like measurement than politics. That might mean we are restricted to counting good or bad things about which few disagree, like infant mortality.
Another difficulty is that some things that make a big difference are hard to measure, like health-gain from an operation, and hard to add to other good things, like clean air, to produce a national total. Or do you have masses of different, incommensurate indicators?
Some of the social indicators the OECD has considered include average years of schooling, gender wage gaps, participation in voluntary groups and suicide rates. But decide for yourself what we should count towards well-being.
In fact, we measure all this already, through the Office for National Statistics (ONS) and others. Volunteering is a great one - so good for us, Evan Davis once joked, that it ought to be made compulsory. Maybe that's what the Big Society is about.
Other research has suggested that what people say improves their well-being - like having children - is at odds with what they report about their happiness in the moment: "Oh no, not the school holidays!"
And even if we could agree what was good for well-being, that doesn't tell us how best to achieve it. That loneliness and isolation are often bad for us might mean that the state should do more - or less, if we think people's relationships thrive when the state gets out of their way and reminds families of their own responsibilities.
So is well-being too woolly, too contentious, too value-laden to be useful? If so, should we give up?
Here's an argument for giving it some thought: even if we can't agree where the responsibility lies for addressing isolation, say, at least we might rethink the weight we attach to it among other priorities. Yes, a policy might help economic growth, but have we properly considered the softer costs and benefits?
On the practical side, there is evidence that children can be taught emotional resilience - lessons derided by some as "happy classes" - but this does indeed improve their subjective sense of well-being.
The simple exercise of introducing feelings to the national accounts, or thinking of them as a legitimate objective of policy, or of stretching the measure of national progress to include things that are not easily monetized, is horribly difficult, but might be a way of reminding ourselves of what we most care about.