What makes us happy? According to the testimonies of the people who describe themselves as contented, you need to be highly educated, female, wealthy, married, self-employed and not middle-aged (tell me about it). Misery, meanwhile, comes from unemployment, low income, divorce and poor health.

Rocket science it is not. Nevertheless, the booming discipline of 'happiness studies' continues to excite controversy. Questions abound over cause and effect — for example, are people happier when they marry, or do happy people marry?

And what exactly do we mean by happiness? That we laugh a lot, feel optimistic and secure in our lives, are serenely calm or deliriously hedonistic? In a recent 2008-09 Gallup poll of national happiness, the United States came fifth, and yet at the same time came 89th out of 151 in terms of 'least worry', and had the fifth-highest stress levels. It is hard to make sense of that: does happiness compensate for stress, or are they ineluctably conjoined?

Besides, is happiness a desirable goal? That might seem obvious (it was to the authors of the US Declaration of Independence) — and it surely seems a better measure of human wealth than conventional economic indices of well-being, such as gross domestic product. But what if a happy nation is a selfish or a profligate one? And who's to say that the inhabitants of Aldous Huxley's Brave New World would not, blissed out by the drug soma, have rated high on the happiness scale?

Money matters

One of the most contested issues for those trying to measure happiness is its relationship to income. Most people would agree that abject poverty is miserable, but it is not clear how the relationship plays out above that unfortunate state. Whereas being female or married are all-or-nothing factors, income is quantitative: if being wealthy makes you happy, does being more wealthy make you more happy?

Contentment — just a (large) payrise away? Credit: Comstock Images / Getty Images

Because most of us are, by definition, not relatively wealthy in our society, we probably feel a glow of self-righteous satisfaction from studies suggesting that there is a 'wealth threshold' above which happiness no longer increases1,2. That fits with intuition: the super-rich do not always strike us as a particularly joyful bunch.

But now Nobel laureate economist Daniel Kahneman and his colleague Angus Deaton at Princeton University, New Jersey, have thrown a cat among the pigeons. In a paper in the Proceedings of the National Academy of Sciences USA3, they use the US data from the Gallup survey to argue that income does continue to impact on our evaluation of 'life satisfaction' as we enter the realm of the rich.

Does this validate the anonymous quip that those who say money can't buy happiness don't know where to shop? Not exactly. Kahneman and Deaton argue that previous discussions on happiness have been muddied by a failure to distinguish a sense of emotional well-being from our life evaluation. The first refers to daily experience: how much we laugh, for example, and how relaxed we feel as we go about our life. The second is a more objective overview: how content we are with our family, job, house, insurance or credit rating. It is not hard to imagine the head of a big corporation feeling good about all this while never cracking a grin.

Vote winner?

The Gallup poll surveyed more than 700,000 US residents, although Kahneman and Deaton jettison about one-quarter of the responses because they appear to be unreliable. From the rest, they deduce that income is more closely correlated with life evaluation than with emotional well-being, and that this correlation persists for all income levels, at least up to an annual income of around US$160,000. Although reported well-being also generally increases with income, this relationship reaches a plateau at an income of around $75,000 a year.

For all their ambiguities, happiness studies are closely monitored by politicians and policy-makers, not least because policies that make people happy seem likely to win votes. The question is what they will make of these findings. Is it better to promote good life evaluation or emotional well-being?

Kahneman and Deaton refrain from taking a position — and the richness and subtlety of their data advise against glib answers. As they imply, any society should wish to improve the lot of people who have poor emotional health and are gloomy about their prospects. But their results, while complicating the previous picture, surely suggest that income (and dare one add, taxation levels) should not be regarded as a relevant happiness dial for those who are well off.

Although some might be determined to extract the conclusion that, as The New York Times once put it, "maybe money does buy happiness after all", there is a stronger case to be made here that better education, secure health provision, lowering of stress, and the nurturing of social and familial relationships offer a far greater dividend of smiles.