The central question for students of well-being is the extent to which people adapt to circumstances. Ten years ago the generally accepted position was that there is considerable hedonic adaptation to life conditions. The effects of circumstances on life satisfaction appeared surprisingly small: the rich were only slightly more satisfied with their lives than the poor, the married were happier than the unmarried but not by much, and neither age nor moderately poor health diminished life satisfaction. Evidence that people adapt — though not completely — to becoming paraplegic or winning the lottery supported the idea of a "hedonic treadmill": we move but we remain in place. The famous "Easterlin paradox" seemed to nail it down: Self-reported life satisfaction has changed very little in prosperous countries over the last fifty years, in spite of large increases in the standard of living.
Hedonic adaptation is a troubling concept, regardless of where you stand on the political spectrum. If you believe that economic growth is the key to increased well-being, the Easterlin paradox is bad news. If you are a compassionate liberal, the finding that the sick and the poor are not very miserable takes wind from your sails. And if you hope to use a measure of well-being to guide social policy you need an index that will pick up permanent effects of good policies on the happiness of the population.
About ten years ago I had an idea that seemed to solve these difficulties: perhaps people's satisfaction with their life is not the right measure of well-being. The idea took shape in discussions with my wife Anne Treisman, who was (and remains) convinced that people are happier in California (or at least Northern California) than in most other places. The evidence showed that Californians are not particularly satisfied with their life, but Anne was unimpressed. She argued that Californians are accustomed to a pleasant life and come to expect more pleasure than the unfortunate residents of other states. Because they have a high standard for what life should be, Californians are not more satisfied than others, although they are actually happier. This idea included a treadmill, but it was not hedonic – it was an aspiration treadmill: happy people have high aspirations.
The aspiration treadmill offered an appealing solution to the puzzles of adaptation: it suggested that measure of life satisfaction underestimate the well-being benefits of life circumstances such as income, marital status or living in California. The hope was that measures of experienced happiness would be more sensitive. I eventually assembled an interdisciplinary team to develop a measure of experienced happiness (Kahneman, Krueger, Schkade, Stone and Schwarz, 2004) and we set out to demonstrate the aspiration treadmill. Over several years we asked substantial samples of women to reconstruct a day of their life in detail. They indicated the feelings they had experienced during each episode, and we computed a measure of experienced happiness: the average quality of affective experience during the day. Our hypothesis was that differences in life circumstances would have more impact on this measure than on life satisfaction. We were so convinced that when we got our first batch of data, comparing teachers in top-rated schools to teachers in inferior schools, we actually misread the results as confirming our hypothesis. In fact, they showed the opposite: the groups of teachers differed more in their work satisfaction than in their affective experience at work. This was the first of many such findings: income, marital status and education all influence experienced happiness less than satisfaction, and we could show that the difference is not a statistical artifact. Measuring experienced happiness turned out to be interesting and useful, but not in the way we had expected. We had simply been wrong.
Experienced happiness, we learned, depends mainly on personality and on the hedonic value of the activities to which people allocate their time. Life circumstances influence the allocation of time, and the hedonic outcome is often mixed: high-income women have more enjoyable activities than the poor, but they also spend more time engaged in work that they do not enjoy; married women spend less time alone, but more time doing tedious chores. Conditions that make people satisfied with their life do not necessarily make them happy.
Social scientists rarely change their minds, although they often adjust their position to accommodate inconvenient facts. But it is rare for a hypothesis to be so thoroughly falsified. Merely adjusting my position would not do; although I still find the idea of an aspiration treadmill attractive, I had to give it up.
To compound the irony, recent findings from the Gallup World Poll raise doubts about the puzzle itself. The most dramatic result is that when the entire range of human living standards is considered, the effects of income on a measure of life satisfaction (the "ladder of life") are not small at all. We had thought income effects are small because we were looking within countries. The GDP differences between countries are enormous, and highly predictive of differences in life satisfaction. In a sample of over 130,000 people from 126 countries, the correlation between the life satisfaction of individuals and the GDP of the country in which they live was over .40 – an exceptionally high value in social science. Humans everywhere, from Norway to Sierra Leone, apparently evaluate their life by a common standard of material prosperity, which changes as GDP increases. The implied conclusion, that citizens of different countries do not adapt to their level of prosperity, flies against everything we thought we knew ten years ago. We have been wrong and now we know it. I suppose this means that there is a science of well-being, even if we are not doing it very well.