How are you feeling today?

With war spreading erratically across the Middle East, Friday 20 March this year doesn’t look a particularly apposite choice for International Day of Happiness.

And the accompanying league tables are inevitably open to question: the World Happiness Index (WHI), sponsored by Gallup, the UN and Oxford University, lists Israel as the eighth happiest of 140 countries, outclassed only by those long-term winners, Finland (top) and other Nordics, the Netherlands - and a new well-being star, Costa Rica.

Meanwhile there has been an unhappy coincidence between declining scores in English-speaking countries and diminishing government enthusiasm for the analysis. The United States is languishing in 23rd place, while the UK has sunk to 29th. Canada, Australia and New Zealand are all out of the top 10. While a UN-sponsored expert group is still doggedly seeking measures of well-being acceptable to a wide range of members, indices like the WHI could hardly be expected to appeal to President Trump.

Feeble economic growth has turned governments’ minds back to output, with happiness (or well-being) seen as requiring interventions unaffordable without growth. At least one UK institute focused on the topic has rebranded itself. It now focuses on “prosperity” rather than “well-being”, while the Office for National Statistics, which was early into the game with a framework for life satisfaction measurement set in 2011, has been scaling back. From publishing 60 well-being measures every quarter, it has cut back to seven.

Move over QUALYs, it’s time for WELLBys

Economists’ efforts to measure national performance go back centuries. But when GDP (and that earlier favourite statistic of national performance, GNP) were “invented” in the 1930s, with Simon Kuznets’ seminal work for the US Department of Commerce, they quickly become, for politicians and journalists, the go-to measures of government achievement.

However, Kuznets himself always maintained that GNP/GDP were measures of activity, not success. And microeconomists have been working for centuries on broader measures of satisfaction, with the concept of social utility taking them quite a long way forward. However, their efforts to find better measures became particularly enthusiastic at the beginning of the new millennium.

Language was a bit of a problem: happiness seemed immeasurable, satisfaction rather limited - while the concept of “well-being” got uncomfortably entangled with other concepts and has been invoked by everyone from foodies to beauty parlours. Still, governments gave airtime (and resources) to the development of new indices, landing most often on “subjective well-being” (SWB) - subjective, because the favoured measure of happiness is a scale on which those surveyed rate their satisfaction with their lives.

It was an admirable effort to move on from the statistics traditionally relied on by macroeconomists to the “lived experience” of the people to whom they were supposed to relate. And it got political traction because an (admittedly limited) amount of analysis of the connection with voting patterns seemed to confirm the reasonable assumption that higher levels of SWB were correlated with electoral support for incumbent political parties. So governments, it seemed, had a survival interest in trying to increase well-being - as well as preventing the use of social media to influence well-being adversely.

This new work had an important influence on policy-making, increasing, just for example, the focus on neglected mental health services, and reinforcing the understanding that employment is good for personal well-being as well as for personal income. But efforts to conduct “well-being assessments” for pieces of legislation have not fared well.

While assessments of QALYs (quality-adjusted years of life) have become well-embedded in government policy decision-making in many advanced economies, efforts to embed WELLBys (Well-Being Adjusted Life Years) have been less successful. In the UK, the Treasury Green Book sets out guidelines, but the promoters of different pieces of legislation are highly selective in their use, and legislators tend to pay even less attention to them than to the better-established “economic impact assessments”. An obvious example is the current Children’s Well-Being and Schools Bill, which has been lumbering its way through Parliament based on no published WELLBy analysis nor data on children’s SWB.

Instead, protagonists for one measure or another tend to select the pieces of data that make their case. The ONS’s 60-indicator dashboard is an example of how broad the subject can become; on the other hand, reducing the scorecard to seven elements tends to create a display of the bleedin’ obvious (ill-health reduces SWB. Who knew?)

Meanwhile, the renewed focus on mental health, one of the big successes of the early development of SWB indicators, has become associated in the UK with the rapid rise in the number of people off work with long-term sickness, of which 40% ascribe this to stress or other mental issues. Policy-makers are struggling to differentiate between different degrees of mental illness across a range of policies, from welfare benefits to extra time for public exams, while clumsy regulation causes resentment.

It’s generally accepted that Covid-19 left a legacy of mental illness, but has it really increased so much? Or was there that much unrevealed mental illness before? Indeed, have we yet properly identified, let alone treated, its most damaging impact, particularly amongst children? And have we meanwhile succeeded in encouraging some teenagers and adults to label the normal stresses of human life, or deviations from some behavioural norm, as debilitating mental illness? And is that increasing or reducing their SWB?

The evidence is in: happiness is a long-term game

What can be done to rescue some important ideas from this mangled history? First of all, the analysis of well-being needs to be seen as addressing so much more than the problem of mental illness, where it must fall to medical practitioners and health policy-makers rather than economists to take on the task of differentiating between the various forms and degrees of illness. But meanwhile, the simplest WELLBy assessment would suggest a focus on children, when improving their SWB will have the longest payback period. Well-being data indicate a marked decline in happiness among young people in the past 15 years, and other factors, such as employment prospects, are clearly also at play.

Secondly, we should concentrate on making better use of the evidence. Those irritating league tables do still provide us with rich data. A key piece of evidence is the damage to SWB from the Covid-19 crisis, from which few western countries have yet fully recovered, even though GDP has finally made it back up over the baseline in even the worst hit. This would be a better framework for Covid-19 inquiries than using hindsight to judicially investigate process details, or to tell us what we surely already knew: that the NHS was ill-prepared, short on capacity and came close to collapse. Whether closing schools to prevent the spread of infection was better than keeping them open to support children’s development is exactly the sort of hard choice that should be subject to SWB analysis, and may be hugely pertinent to the handling of future crises.

Another important indication, more difficult to assess, is that Nordic countries fare well because their citizens have trust in their institutions and in their safety net of generous and effective welfare. It seems plain that a decline of faith in institutions has damaged well-being in a number of western economies, but the welfare question is more difficult, because in some (like Britain) the sheer scale of welfare systems have become a matter of contention, with increasing public dissatisfaction at their cost. But if recent attempts to cut costs have taught government anything, it should be that a robust well-being impact analysis is essential to overcoming the political difficulty of controlling spending.

The 2026 World Happiness Report is focused on another area where well-being analysis is clearly essential to good policy-making: the impact of social media. Again, it is the impact on children that most needs to be grounded in analysis. The evidence show-cased in the report gives a broadly negative picture, but also useful evidence of the extent to which the impact on well-being varies not only by age-group or gender but also socio-economically. Whether restrictions contribute more or less to well-being than open access to the digital world is a question that should be settled by data rather than by trigger-happy politics.

Of course, there’s a limit to the value of this approach. Government finance ministries are understandably wary, particularly in straitened times. The diversity of factors that influence SWB creates a long list of possible interventions (e.g., to create green spaces, community networks, gyms, therapies) that without the discipline of rigorous comparison risk bankrupting government. And well-being policy shares with utilitarianism the danger of curtailing freedom, taking society along a slippery slope towards the compulsory prescription of happy pills for all.

Moreover, some hugely important well-being factors, such as a sense of security, are only likely to be picked up in the indices when it is too late to do anything about them. Countries under threat may display high morale, bolstering a sense of well-being. But most war zones, unsurprisingly, show low levels of SWB. However the levels of defence spending that might protect against becoming a war zone have to be taken years and even decades in advance.

Too great a reliance on well-being indices could easily become yet another excuse for spending on short-term issues rather than grasping the long-term danger. Other long-term factors, such as social polarisation, and a loss of social trust, do already show up in the indices but are also not susceptible to a quick fix. But for those in politics who still try to maintain long sight, they should still provide important reminders of at least part of the bigger picture.