Yesterday, I went to the British Academy to hear Richard Easterlin, the father of happiness economics, present his latest thinking, together with Andrew Oswald of Warwick University. Sir Gus O’Donnell, head of the civil service, was also there – and made an interesting interjection.
Easterlin asked the provocative question: does higher income raise happiness in poorer countries? His answer was ‘no, as far as the evidence goes’. He showed graph after graph of transition economies where the income has been rising sharply over the last decade, while the happiness levels remained flat. He focused on China, a country “where the rate of economic growth has been completely unprecedented, at almost 10% a year for the last decade. If any country would show an improvement in happiness, it would be China.” But it doesn’t. Another flat line. The only countries which showed noticeable drops and rises in happiness levels, as far as I could tell, where post-Communist countries, whose happiness levels showed a clear (and understandable) drop after the collapse of communism, and then a steady rise after that, only to flatten out again.
This famous flattening of average happiness levels despite rises in income has been called the Easterlin Paradox, and is perhaps the single most influential graph for happiness economics. It is used, over and over again, as evidence that governments should not be focusing on raising income, but instead on raising happiness.
I asked Easterlin: do any policies have any clear impact on national happiness levels? Perhaps the happiness flatlines we see in country after country is evidence not that we’re pursuing the wrong policies, but simply that our daily happiness levels are not very sensitive to major policy changes. Think about all the different political, economic and cultural changes over the last 50 years in the UK, and yet our happiness levels remain flat. Why is that? I suggest it’s because the measurement technique – asking people to rate their happiness between one and ten – simply isn’t good enough to pick up changes in quality of life over time. We adapt to our situation, and except in moments of extreme crisis, we always say ‘oh, about a seven’.
What do happiness economists expect? Do they think that, if governments pursue the right policies, the public will go from a seven, to an eight, until eventually, after say 30 years, we will all be shouting ‘Ten!’ before ascending in rapture unto heaven? Of course, given such a bounded numerical scale, people are going to say ‘about a seven’, even if their lives have actually got better or worse over time. We forget the bad times, and we also forget the good times. Our daily well-being is probably protected by our forgetfulness and our ability to adapt.
Happiness economists try to get around this by using country comparisons. ‘Look’, said both Oswald and Easterlin, ‘how Scandinavian countries are typically happier than Anglo-Saxon countries. This is because they spend more on health, education and unemployment benefits. If we did the same, we’d be happier.’ I’m paraphrasing, but that’s basically what they both argued. Now I’m personally all for higher education and health spending. But that sort of cross-country comparison completely ignores cultural differences.
To be convinced, I’d like to see examples within a particular country where particular policies have led to a clear rise or fall in national happiness levels. Do such examples exist, I asked. “Yes”, Easterlin replied. “There are clear links between employment and happiness levels in countries. Unemployment in the US has dropped markedly in the last three years, and happiness levels in 2010 were at their lowest for many years.”
At that point, the head of the civil service, Sir Gus O’Donnell, who was listening attentively in the audience, joined the debate. He said: “One of the things we’re trying to figure out is the adaptation effects. There’s a new paper out by Angus Deaton, which looks at the effect of the recession of US happiness levels, and it shows that happiness levels are already back to their pre-crisis levels, despite unemployment still being much higher than it was. There’s even evidence that people adapt quite quickly to traumas like losing an arm. So what does that mean for public policy?”
It’s worth having a look at the paper O’Donnell mentioned (it’s also written up in the new issue of The Economist) because it is quite damning for happiness economics. It finds, for example, that subjective well-being measurements seem only very slightly correlated with the doubling of unemployment in the US, while there are clear spikes on Valentine’s Day and Christmas. Deaton notes that subjective well-being levels seem to be very correlated with stock market levels, and that perhaps they are driven by the same short-term sentiment factors, like headlines, holidays or the weather, rather than serious policy changes.
He writes: “While it is conceivable that, as is sometimes argued for the stock market, the SWB measures are giving an accurate take on expected future well-being, it seems more plausible that, like the stock market, they have actually very little to do with well-being. As McFadden suggests in comments on this paper, we may be looking at “cognitive bubbles” that are essentially irrelevant for any concept of well-being that we care about…[SWB measurements] still have a long way to go in establishing themselves as good time-series monitors for the aggregate economy. In a world of bread and circuses, measures like happiness that are sensitive to short-term ephemera, and that are affected more by the arrival of St Valentine’s Day than to a doubling of unemployment, are measures that pick up the circuses but miss the bread.”
Strong words – and interesting that O’Donnell should have picked up on them. It shows, I would suggest, the Civil Service’s own queries about the mission they have been set by the Coalition Government – but it also shows how seriously O’Donnell is taking this mission, and his own interest in it. I would suggest that empirical studies of what factors lead to flourishing can and should influence policies. I’m just not convinced by broader arguments based on national happiness levels. The measuring instrument is simply too blunt to be of any real use to policy makers.