Unpacking Happiness

A central claim of this project is that the purpose of government, stated explicitly, should be to maximise the sustainable wellbeing of its citizens. That is an argument that requires unpacking before it can be properly defended, because it rests on a prior set of questions that are both more contested and more interesting than the claim itself. What is happiness? Is it the kind of thing that can be pursued collectively, at the level of the nation, without distorting it beyond recognition? How do you measure it? And what, empirically, produces it?

This essay works through all four of these questions. But first, an important point about terminology - much of the discussion in this discipline talks about happiness. This is a more accessible and meaningful term for most people than wellbeing. There is a qualitative difference between the two concepts though. Happiness as it is commonly understood implies an emotional state of joy, or at least contentment. Wellbeing is more appropriate to describe the broad aspiration of a life that best meets a person’s desires, which is not precisely the same thing - wellbeing is entirely possible without happiness, if, for instance, a person defines their wellbeing in terms of good physical health, or prosperity. However, for simplicity, I am going to largely use the two words interchangeably throughout this project.

What is happiness?

The word happiness is, in some ways, an obstacle. “Happiness” sounds modest — even slight — next to grander-sounding objectives like “national prosperity” or “self-actualisation.” But the philosophical tradition that has grappled with it for two and a half millennia takes a rather different view. To philosophers from Aristotle to the present, the question of what makes human life go well is arguably the most fundamental question in all of ethics. The difficulty is that they have not agreed on the answer.

Aristotle’s answer was eudaimonia — usually translated as “happiness” or “flourishing,” though neither fully captures the original sentiment. For Aristotle, eudaimonia is not a feeling but a mode of living: the full exercise of distinctively human capacities — reason, virtue, deliberate action toward worthy ends. A life goes well when it is lived in accordance with what humans are actually for, not merely when it feels pleasant. This is a demanding standard. It means that a life of passive contentment, with all material needs met and no friction whatsoever, might still fall short. A human life is well-lived through what it does, not only through what it experiences.

Epicurus offered a subtly different version. His goal was ataraxia — tranquillity, freedom from anxiety and disturbance — rather than stimulation and excitement. The pleasant life, for Epicurus, is not one saturated with pleasures but one in which pain and fear are absent. Friendship, modest living, and philosophical reflection were its primary constituents. There isn’t a great philosophical distance between Epicurus and Aristotle but the emphasis is meaningful.

The Stoics pushed further still, arguing that external circumstances — wealth, health, social standing — are neither good nor bad in themselves, only preferred or not preferred. The only genuine good is virtue; the contented slave may be freer, in a morally significant sense, than the tormented emperor. It is a demanding position, and for policy purposes, the most uncomfortable: if the Stoics are right, the contribution that governance can make to human flourishing is marginal at best. Government can provide the material conditions for virtue; it cannot produce the virtue itself.

Two centuries ago, the philosophical conversation began to take a more empirical turn. Maslow’s hierarchy of needs proposed a layered structure in which basic physical and safety needs, once satisfied, give way to social and esteem needs, and those to the aspiration toward self-actualisation — the realisation of one’s individual potential. The structure is contested in its details but contains an important insight: the question of what matters to human wellbeing is not static. As one set of needs is met, others become salient. This is not a trivial observation for policy: wellbeing exists in a context, and that context is important.

Martin Seligman’s PERMA framework — positive emotion, engagement, relationships, meaning, and accomplishment — is the most influential recent attempt to systematise the components of flourishing. It is deliberately pluralist: no single element is privileged, and a rich human life typically requires all five. It is also more practically useful than most philosophical accounts, because its components are at least partially observable and therefore potentially measurable.

One further perspective is important to briefly flag, and it is simultaneously liberating and unsettling for anyone who thinks government can engineer happiness: set point theory. The work of Lykken and Tellegen in the 1990s, replicated and extended since, suggests that roughly half of an individual’s happiness level is genetically determined — that people adapt to changed circumstances, returning over time to something like a baseline level of subjective wellbeing, and that the long-run effect of most life events on happiness is smaller than intuition suggests. Lottery winners are not much happier a year after winning. People adapt to disability, bereavement, and hardship in ways that are both remarkable and somewhat humbling to anyone in the business of improving lives.

What should we make of all this? For present purposes, the answer is: not too much. Each tradition is capturing something real. But across all the different perspectives, a common thread runs: what makes a human life go well is not simply the accumulation of pleasant experiences, but something more enduring about the conditions under which people are able to live with purpose, freedom, and connection.

For the purposes of this project — and for the purposes of policy — I would suggest that we do not need to resolve the philosophical debate. All we really need is a working concept of what we are trying to produce, and a way of recognising whether we are succeeding. I have argued in the introductory essay that for me, happiness is effectively whatever the individual decides it is. Happiness is, by this definition, the thing that everyone individually aspires to, and hence it essentially defines itself. Ultimately, for practical purposes, happiness is necessarily defined by what we can measure and really nothing else. Happiness is whatever the measurement tool says it is. Measurement by this logic becomes very consequential, but it is a surprisingly difficult challenge, which I will come to later.

Happiness as a collective objective

Before we tackle measurement, it is worth touching on what it means to pursue happiness collectively, as a society. The step from individual happiness to collective happiness, and what it means for governance and policy, is not automatic, and it is worth being honest about the difficulties.

The foundational objection is self-interest: individuals pursue their own wellbeing, often at the expense of others. The natural state of human social life, Thomas Hobbes argued with bluntness, is competition for scarce resources — and without authoritative governance, that competition produces a war of all against all in which life is “solitary, poor, nasty, brutish, and short.” This is the condition from which the social contract is supposed to rescue us: individuals surrender some freedom to a governing authority in exchange for the security and predictability that makes flourishing possible.

Enlightened self-interest takes this further. In a well-functioning society, the wellbeing of others is not in zero-sum conflict with your own. Social trust, functioning institutions, the public goods that no individual can provide for themselves — all of these make the individual better off, not just the collective. The argument for caring about others’ wellbeing is not purely altruistic; it is predominantly self-interested.

A fundamental problem always remains though: the free-rider. These are the individuals who are so nakedly self-interested that they are happy to take advantage of the civilisational benefits, while acting outside the norms and expectations of civilisation so as to benefit themselves. An ugly strain of this mindset is one of the central organising concepts of the manosphere: it advocates such behaviour as smart, and sees the mainstream that accepts the constraints of civil society as naive. Shades of this belief run through many sections of society though, sometimes through a lack of education, sometimes through sheer self-regard. It can often be seen in law and order debates, and in political vilification of the welfare state. It is, unfortunately, a persistent background frequency that seemingly runs through all societies.

If we accept though, that there is a broad consensus that the role of Government is to maximise collective wellbeing, we need to grapple with the detail of what this actually means.

Utilitarianism provides the simplest formal formulation: maximise the aggregate. Add up individual wellbeing across the population and find the policy that produces the greatest sum. It is philosophically elegant and practically tractable, which is why it became the dominant framework in welfare economics. It is also morally dangerous, for reasons that are by now well-rehearsed: it permits the sacrifice of minorities if the aggregate gain to the majority is large enough. The utilitarian calculus can, in principle, justify almost anything, provided the arithmetic works out. This is not an abstract criticism; it is the intellectual foundation that — pursued to its extreme and drained of any rights constraint — underpins fascist claims to act in the name of a collective happiness that absorbs and supersedes the individual. And it is a philosophy that remains alive: China today can be seen almost as an experiment in pure utilitarianism.

Simple utilitarianism is also unable to grapple with a fundamental temporal problem: over what time period is the aggregate to be maximised. Is it the current population, or do future citizens matter. Even within the current population, how much weight should be given to the elderly maximising their enjoyment of their last days, against the newborn with a life of needs and desires ahead of them.

Third, there is a foundational prerequisite that can’t be lost sight of: safety and security. This is a quality of life that isn’t highly valued until it is lost. Governments need to ensure that they are in a position to protect their citizens both from each other, and from outside forces. Diversion of resources away from immediate material needs to fund security has a welfare cost that is a potential deadweight loss for aggregate utility in the present. But it is of immeasurable benefit when unrest and disruption threaten the foundations of society.

My belief is that the resolution of these objections is the three-constraint formulation set out in this project’s welcome essay: the objective of government should be to maximise sustainable societal wellbeing, subject to a floor of minimum human rights, and the precondition of physical security. Sustainability disciplines the inter-generational effects; the rights floor prevents the utilitarian slide toward the sacrifice of minorities; physical security identifies the prerequisite on which everything else depends.

Measuring happiness

Returning now to the question of measurement, it should be self-evident, but to be clear: if governments are to pursue wellbeing as an explicit objective, they need to measure it.

For some, there is an intuitive objection to this exercise, that wellbeing is too subjective to be measured reliably. However, this confuses two different things. Subjectivity is the property of being experienced from the inside; unreliability is a property of measurement methods. Subjective experience can be consistently and reliably reported, provided the methodology is sound and the question is well-designed. The fact that only you can fully know what your life feels like does not mean your report of how it is going is meaningless.

A significant institutional endorsement of this view came in 2013, when the OECD published its Guidelines for Measuring Subjective Wellbeing — a serious, technical document that treated self-reported wellbeing as a legitimate complement to objective economic and social indicators, and provided recommendations for how national statistical agencies should collect it. The OECD’s endorsement matters not because it settles the methodological debate, but because it signalled that subjective wellbeing measurement has moved from the margins of welfare economics to its mainstream.

The alternative approach — recommended by the Stiglitz-Sen-Fitoussi commission and influential in European statistical practice — is to build a dashboard of multiple indicators: GDP supplemented by measures of inequality, social connection, environmental quality, health, and so on. The instinct behind this is sound. GDP is clearly inadequate as a single measure of societal success. But the multi-indicator approach has a fundamental structural problem: every choice it makes — which indicators to include, how to weight them, what benchmarks to use — embeds a value judgement. Researchers make these choices, and their choices reflect their values. This is not a criticism of their integrity, it is a structural feature of the method. The question of how to weight longevity against freedom, or material comfort against equality, is precisely the kind of normative question that democratic governance is supposed to answer through deliberation and choice. When technical bodies pre-emptively answer it instead, they are doing politics under the cover of statistics.

The measure that avoids this problem most cleanly is the Cantril ladder: a zero-to-ten scale on which respondents are asked to place themselves, with “ten” representing the best possible life for them and “zero” the worst. It is the primary measure used by the World Happiness Report, which has been tracking it annually since 2012. Its virtues are simplicity, cross-cultural applicability, and longitudinal consistency. It is also sufficiently intuitive that its results are legible to a non-specialist audience. This matters, because a measure that serves only researchers serves the governance objective poorly.

But I see the Cantril ladder as having two related structural limitations worth being explicit about.

The first I think of as a boundedness problem. Unlike GDP, which has no upper limit, the Cantril scale has a fixed ceiling: ten. The question is anchored to the respondent’s own concept of the best possible life — not to some objective standard of human flourishing. This means that if everyone’s circumstances genuinely improved, but their internal reference point for “best possible life” moved upward in proportion, the measured score could remain flat. The measure tracks relative self-assessment, not absolute flourishing. This is not a data quality problem; it is a structural property of the instrument. It means that observed changes in happiness scores — especially modest ones — need to be interpreted with care.

That said, I don’t believe that this diminishes the utility of the Cantril ladder as the defining measure of government success. My argument is that the role of government is to maximise sustainable subjective wellbeing. The possibility that the score has declined because respondents’ assessment of the best possible life has increased doesn’t diminish the meaningfulness of the result. Government should be aiming to increase wellbeing faster than respondents’ subjective assessment of the best possible life increases, and that is what the ladder measures. The aim is to reduce the size of the gap. To the extent that a government has shifted subjective perceptions of the best possible life by its policies, that may be a good thing, but it is something different to what I am arguing should be the government’s defining objective.

The second limitation is what might be called the political appeal problem. Consider the asymmetry between GDP and the Cantril ladder as political objects. GDP, in practice, almost always goes up — the political debate is never whether it grew, but whether it grew fast enough. A government defending 1.5% growth against an opposition promising more is on difficult but manageable terrain. A government defending a decline in the national happiness score faces a categorically different problem: the measure has gone in the wrong direction. Here, Kahneman and Tversky’s prospect theory becomes directly relevant. Their foundational finding is that losses are felt approximately twice as intensely as equivalent gains feel good — a cognitive asymmetry so robust it has been replicated across cultures, contexts, and states. Applied to political psychology: a one-point decline in a published happiness index would carry roughly twice the political pain that a one-point rise carries in political benefit. The measure’s boundedness — its fixed ceiling at ten — means it oscillates in a way that GDP does not. For politicians already hypersensitive to bad news, this is a structural deterrent to adoption quite separate from any questions about the measure’s validity. It is inherently threatening as a mainstream measure of political performance. This is not a reason to abandon it — but it explains something about the institutional resistance the measure has encountered, and it argues for making the political case for these measures as explicitly as the methodological one.

While I lean to the view that the Cantril ladder methodology is meaningful, and a pragmatic solution to the desire to measure wellbeing, the subjectivity critique isn’t without merit. Self-reported happiness is filtered through expectations, cultural norms, social comparison effects, and a lifetime of adaptation. It tells us something real and something important. But it cannot tell us whether people are genuinely flourishing at some more fundamental level, or whether they have simply adjusted their expectations to match their circumstances.

Research into biological correlates of wellbeing exists and points in interesting directions. Neuroscience has identified left prefrontal cortical asymmetry — greater activation in the left versus right prefrontal cortex — as a consistent marker of positive affect, detectable via EEG. Structural brain imaging has found associations between grey matter density in specific regions involved in self-referential processing and self-reported happiness scores. Heart rate variability, which reflects parasympathetic nervous system activity, has emerged as a more accessible physiological proxy: higher variability is consistently associated with better emotional regulation and higher subjective wellbeing, and it is now routinely measured by consumer wearables. These are genuine findings, not speculation, and they point to ways to objectively measure wellbeing.

However, they remain laboratory-scale observations rather than a measurement system. The more accessible proxies — heart rate variability in particular — measure components of positive affect and stress regulation, not flourishing in any fuller sense. And here the philosophical problem may ultimately prove as hard as the technical one: even a robust biological correlate of feeling good would not necessarily capture for instance what Aristotle meant by eudaimonia — a person can be neurochemically calm and still be living a life of shallow purpose.

Until a measure emerges that is both biologically grounded and philosophically adequate, self-reported wellbeing data is what we have. It provides a valid foundation to work from, but it is worth being clear-eyed about what that means for the conclusions we can draw from it.

Explaining happiness

Among the more important of those conclusions is the question as to what explains collective happiness. Measuring wellbeing is only the beginning. The more policy-relevant question is what drives the variation in national happiness scores across countries and over time. That is where the roadmap for decision-making lives.

The World Happiness Report’s approach identifies six explanatory variables: GDP per capita; social support; healthy life expectancy; freedom to make life choices; generosity, and; perceptions of corruption. Together, these account for a substantial proportion of the variation in national happiness scores, with robust statistical validity. But they are better understood as intermediate variables — things that correlate with happiness rather than primary causes. “Perceptions of corruption” is a measure of institutional quality. “Social support” is a measure of social connection. Each of them raises a further question: what produces them?

This is the space where I see both the greatest gap, and the greatest need. As discussed at length elsewhere in this collection of essays, it seems clear that the world is failing to deliver the flourishing that I would argue is the central defining purpose of a society. Figuring out, with precision, what enables that flourishing, and crafting the necessary policies to foster it, must be central to the future direction of Government.

I would make two observations on this gap in our understanding.

First, I wonder whether too little attention has been paid to culture. Future essays will dive deeply into the question of what is national culture, and why it might be critically important. For the moment though, it is worth observing that culture is like a binding agent in which society exists. If the values and behaviours that infuse a society, and which define its state, are not conducive to individual and collective flourishing, any policies overlaid on that culture are likely to have little traction.

Second, this essay earlier referenced Maslow’s hierarchy of needs. At the level of nations, it is fairly obvious that the citizens of a country in a state of civil war have a different set of needs to those of a rich, stable, democracy. Yet this isn’t a characteristic of current unified models of national happiness outcomes. It is worth flagging here that this is possibly more an issue for wealthy countries than for developing countries. It is clear that measures like per capita GDP work across a wide range of national circumstances - but western countries are increasingly becoming post-materialist. At this point of their development, their failing may be in not recognising Maslow’s concept of self-actualisation as potentially being the essential driving force for happiness. This doesn’t diminish the importance of pinpointing the root explanations of happiness in developing countries, where the majority of the world’s population live. Rather, it is simply saying that explanations of wellbeing need to be sharper at every level of development.

More broadly, there is a point to be made here about the behaviour of Government that ultimately weaves its way through all of the content in this project: people deserve to be governed with a focus on what the evidence suggests works, rather than on what ideology or electoral calculation recommends. The world has no shortage of governments confidently pursuing policies designed to make their citizens better off, on the basis of theories that have not survived empirical scrutiny. A commitment to wellbeing as the governing objective ought to carry with it a commitment to following the evidence about how wellbeing is actually achieved — including when that evidence is uncomfortable. I readily acknowledge here that there is a troubling deficit between the evidence that is needed and what is available. However, that simply points to the need for society to systematically prioritise a strategic agenda of moving from clear objectives, to evidence based policy formulation, to implementation.

Where Does This Leave Us?

So, to recap these four topics: Philosophically, the traditions disagree at the margins but converge on the view that human flourishing is more than material comfort — that it involves the exercise of distinctively human capacities in conditions of freedom, connection, and purpose. I would argue an important take-away from this essay though is simply that the debate as to what happiness is, is academic - in practice it is defined in functional terms by what it is possible to measure. Politically, extending that concept from the individual to the collective requires a careful formulation that captures the best of utilitarianism while protecting against its pathologies: short-termism, the tyranny of the majority, and a strong foundation of public safety and security. Measurement is, I would argue, more tractable than sceptics assert, though the available tools do carry real limitations, and our current best measures are probably not the best measures possible. And then the empirical picture: this is the space where real work is required, to not just identify correlations, but to truly understand how societies at all stages of development can lift the life satisfaction of their citizens.

← Back to Home