The pursuit of happiness is a right enshrined in the Declaration of Independence. So an increasingly important question for economists, psychologists and decision-makers is the role that happiness plays in society and how to increase it.

In the past, economists have more or less ignored this subject, assuming that happiness correlates with macroeconomic conditions such as gross domestic product per capita. But this link has come under fire in recent years.

Various surveys indicate that perceived life satisfaction is higher in poorer countries such as Brazil, Costa Rica, and Panama than in North America. That pulls the rug from this kind of thinking and has profound implications for the kinds of decisions that economist and politicians must make about the future.

But working out exactly what factors affect happiness is not easy. The data is difficult to gather and the statistics are hard to manage.

One of the best sources of data is the General Social Survey, a set of questions about attitudes in the U.S. which has been carried out since 1972. Consequently, it now provides a substantial database for sociologists, demographers and economists studying changes in the way people think and feel in the U.S.

In particular, the survey asks: “Taken all together, how would you say things are these days—would you say that you are very happy, pretty happy, or not so happy?” Some 32,000 people have answered this question over the years, providing a rich source of data.

Today, Teng Guo and Lingyi Hu, who do not give their affiliation, provide a detailed statistical analysis of this data to try and tease apart the factors that determine happiness in the U.S.

They break their analysis into two parts. The first looks at how happiness correlates with personal conditions such as age, health, marital status and personal income and so on. The second looks at the correlation between happiness and macroeconomic indicators such as the rate of inflation and the GDP per capita.

The results are interesting. The biggest personal factor in determining happiness is health. Healthy people are about 20 percent happier than average while unhealthy people are about 8.25 percent more unhappy.

Next comes marriage. Married individuals are about 10 percent happier than people who have never been married.

Personal income plays a smaller role. In general, however, people with higher incomes are happier, with the people in the highest income bracket about 3.5 percent happier than average.

This may help to explain one of the analysis’ more curious findings: that having children reduce happiness. On average, each child reduces happiness by about 0.24 percent. Guo and Hu say this is probably because the survey is biased toward poorer families with less disposable income. Children eat up spending money and this increases hardship.

By contrast, the links with macreconomics factors are much harder to spot, say Guo and Hu. For example, they find it hard to identify a link between happiness and GDP or change in GDP. In fact they say the data indicates there is “no noteworthy connection between the two variables statistically.”

However, inflation does seem to reduce happiness. Guo and Hu say the indicates that a 1 percent increase in inflation reduces national happiness levels by about 3.1 percent. “The intuition behind this result is straightforward, an increase in inflation results in a decrease in purchasing power due to a rise in prices.”

So what to make of this study? On the face of it, the results seem to indicate a clear path for decision makers: the best way to make the nation happy is to improve people’s health.

But Guo and Hu point out that there are significant problems with data of this kind, not least of which is the mechanism of cause and effect. Does being healthy make people feel happier or do happy people tend to stay healthier? Does increasing income make people happier or do happier people tend to make more money?

Nobody yet knows the answers to these kinds of questions. And that makes it hard to make policy decisions based on this kind of evidence. (On the other hand, lack of evidence has never stopped policy makers in the past.)

One thing that policymakers ought to be able to agree on is that more work is urgently needed: the happiness of nations is surely too important to be left to the random forces of chance or to the flawed decision-making processes of politics.

Ref: arxiv.org/abs/1112.5802: Economic Determinants of Happiness: Evidence from the US General Social Survey