• Governments worldwide should measure and monitor well-being using national surveys.

  • An internationally accepted well-being statistic should be developed and communicated to the public.

  • Public debate around this statistic will prompt policy-makers to maximize well-being over economic growth.

In a speech in May, British prime minister David Cameron remarked that a programme under way at the UK Office for National Statistics (ONS) to collect social data and develop decision-making rules based on well-being “may be the most quietly radical thing this government is doing”. The riots in the United Kingdom last month — an expression of social dissatisfaction that was more than a response to economic conditions — highlight the importance of this approach.

Governments in France, Italy, Germany, Spain, the United States, Australia, China and Ecuador are also taking steps to measure quality of life as well as economic growth. The European Commission and the Organisation for Economic Co-operation and Development are harmonizing these efforts. And the General Assembly of the United Nations has called upon member states to place greater emphasis on measures of well-being in their evaluations of social and economic development. My colleagues and I at the New Economics Foundation (NEF) in London have been arguing for such a focus since 2001; we are now working with the UK government and the European Commission to ensure that these efforts are translated into practical implications for public policy.

Family support in Brazil bolsters well-being even when economic resources are limited. Credit: E. MARTINO/PANOS

This interest in well-being — and its subjective measurement — is good news. Economic growth is just one of many tools for bringing about good lives. Political decisions involve trade-offs — between, say, fostering economic growth and stable communities, or agreeable urban landscapes. The traditional focus on gross domestic product (GDP) as a target biases these decisions. The result is lower levels of public well-being than could be the case if people's quality of life was the priority. As economic activity places a greater strain on the environment than many other routes to happiness — such as spending time with one's family — this bias is also bad for sustainability (see 'Good lives needn't cost the Earth').

There are two key challenges for researchers, politicians and policy-makers: first, to gather and interpret new data, so as to create a much fuller science of well-being to rival traditional economics; and second, to create public understanding of some headline measure of well-being and of the role of policy in influencing it, in order to create the political will to use the new science.

The evidence

Well-being is variously defined. Psychologists see it as 'good functioning' or the meeting of psychological needs1, an approach that emphasizes relationships, autonomy, competence and purpose. Economists use more abstract terms such as 'happiness' or 'utility'.

Social surveys over several decades have shown that economic and social policies affect aspects of well-being, however it is defined. Income correlates with well-being, but only up to a certain level, which varies between countries. In the United States, for example, earnings above US$75,000 don't add much more happiness2. Studies also reveal that loss of income is more damaging than a gain is beneficial, and unemployment is more damaging to well-being than is the consequent loss of income. Casual employment is bad for well-being, but self-employment is good, at least for those earning a decent income. Commuting is bad.

Moderate economic growth is good for well-being, but the disruption to lives that is associated with a rapid boom is bad. Inflation is harmful — but a 1% rise in inflation hurts less on average than a 1% rise in unemployment. Fluctuating rates of inflation are worse than stable ones. High levels of short-term debt are associated with low values of well-being. Effective democratic structures, social trust, social networks and time spent socializing are all important for well-being, as are access to parks and views of greenery for city-dwellers.

Figure 1
figure 1


Levels of well-being vary sharply between countries. Sub-Saharan Africa and Eastern European countries score badly. Scandinavia and Latin America do well. In general, there is a positive association between social equality and well-being, although there are exceptions: inequality is high in Latin America, for example, but family support plays a strong role in social security3.

So an economic policy driven by what we already know about well-being would place more emphasis than do current economically centred policies on securing full employment, on stability, and on preserving and strengthening communities and ensuring a reasonable level of income for all. Governments have not yet made those changes. What will cause them to do so?

A single measure

A headline statistic that captures well-being — a measure that voters recognize and can use to hold politicians to account, like the unemployment or economic growth figures — could make the difference. Unless voters care, politicians won't care. And voters won't care without a number — a measure that tells them how things have been going and might be projected to the future. The GDP statistic has had this effect since it was invented in 1934.

The Stiglitz commission, set up in 2008 by French president Nicolas Sarkozy, has laid some of the groundwork for developing a measure of well-being4, but there is more to do. The measure must be able to move and be seen to move over time, in the way that GDP figures do. It must be in a format that allows international comparisons, perhaps accounting for intercultural differences by focusing on movements rather than absolute values.

Most important, it must link well-being to social and economic conditions — and that will require further research. The survey results published are rarely linked to particular policy decisions. And if they are not 'official', the statistics may go unnoticed beyond a given field. The causal factors underlying correlations between happiness and social conditions are often poorly understood, the variables are difficult to control for and the results can seem inconsistent between studies. Large-scale studies that investigate many potential drivers of well-being and also include adequate controls are scarce.

The United Kingdom is leading the way to better measurement with an annual survey by the ONS (on which the NEF has advised) that will probe the well-being and circumstances of 200,000 people. Since April, the ONS has included in its Integrated Household Survey four subjective questions: how satisfied people are with their lives; how happy they were yesterday; how anxious they were yesterday; and how worthwhile they think the things they do are. The survey also asks a large number of questions about objective circumstances that are influenced by government policy — for example, health, housing, education, household income and employment patterns and benefit entitlements.

With such information it will become possible to trace statistical relationships and thereby to assess the relative impact of various policies on well-being. Well-being will then become the common currency for a new form of cost–benefit analysis, which will not replace political judgement and bargaining, but should inform it. Policy-makers will be able to ask, for example, what have been the relative impacts of steps to reduce unemployment in an area and steps to preserve the environment? Have attempts to increase community cohesion and increase economic activity worked? Have the public-health measures in one city been more effective at increasing well-being than those in another?

Theory in action

Our recommendation to the ONS, and to other governments, is that the main measure publicized each year should be based on perceived levels of well-being rather than the drivers of it. That is, it should be based on the responses to the four subjective questions asked in its household survey, rather than on the objective questions and the resulting data. It should take the form of a percentage of the population above a certain threshold level of well-being — the percentage that is flourishing.

The analytical work identifying the drivers of this headline measure must also be presented, in a form that can be picked up by the media and can inform a public debate about which policies to pursue. Voters will come to see that movements in the measure — like those of GDP or inflation or carbon emissions — are the responsibility of politicians. Politicians will then have to take the well-being measures seriously, and act on the advice of the analysts.

Such discussion must not be conducted behind closed ministry doors. The statistical authorities should make the data publicly available at a level of granularity (including geographical detail) useful to a wide range of researchers. To probe causality, researchers will need funding for field experiments and qualitative work. And somehow, politicians and officials will need to be made to listen to the results even when they do not confirm existing views.

All of this is possible, but it will not happen automatically. We must continue to push.