How Much of Happiness Is Genetic?

If you have read anything about happiness in the last twenty years, you have probably seen the pie chart. Half of it is labelled genes, a thin slice of 10% is life circumstances such as income, marriage or where you live, and the remaining 40% is intentional activity: the things you choose to do.

The message is encouraging. Your genes set a baseline, your circumstances barely matter, and a large share of your happiness is in your hands. The chart comes from a 2005 paper by Sonja Lyubomirsky, Kennon Sheldon and David Schkade, and by 2019 that paper had been cited more than 3,000 times.

The trouble is that almost every part of the chart turns out to be shakier than it looks. To see why, it helps to start with where the genetic number came from.

What twin studies found

In 1996, David Lykken and Auke Tellegen published a short paper with a provocative title: 'Happiness is a stochastic phenomenon'. They had measured well-being in several thousand middle-aged twins drawn from birth records, using a standard personality questionnaire.

Twin studies rely on a simple comparison. Identical twins share essentially all their genes, while fraternal twins share about half. If identical twins resemble each other more in some trait than fraternal twins do, that extra similarity is taken as a sign of genetic influence.

Lykken and Tellegen found that between 44% and 52% of the differences in well-being between people were associated with genetic variation. Meanwhile, socioeconomic status, education, family income, marital status and religious commitment each accounted for no more than about 3%. When they retested smaller groups of twins after four and a half and ten years, they estimated that the heritability of the stable part of well-being, the part that persists over time, approached 80%.

Later research broadly agrees that genes matter, but gives lower numbers for everyday well-being. In 2015, Meike Bartels reviewed 30 twin and family studies. The individual estimates varied enormously, from 0% to 64%. Pooling the most comparable ones, the average heritability was 36% for well-being, based on nearly 56,000 people, and 32% for satisfaction with life, based on nearly 48,000.

What heritability does not mean

This is where most popular summaries go wrong. Heritability is a statement about a population, not about a person. A heritability of 36% does not mean that 36% of your happiness comes from your genes. It means that, in the group studied, about 36% of the differences between people were associated with genetic differences between them.

That distinction has real consequences. Heritability depends on how varied the environments in the sample were: if everyone lives in similar conditions, genes will account for more of the remaining differences. It says nothing about how much a person could change if their circumstances or habits changed. And genes rarely act alone. A cheerful temperament may lead you to seek out friends, and friends may in turn make you happier, so what shows up as a genetic effect may partly work through your choices and surroundings.

Why the pie chart falls apart

In 2020, Nicholas Brown and Julia Rohrer published a detailed critique of the happiness pie in the Journal of Happiness Studies. They found that the 50% figure had been derived from three studies in a way that was not clearly specified, one of which was Lykken and Tellegen's. The 10% for circumstances came from older estimates, and the 40% was simply what was left over.

Their central objections were conceptual. First, splitting variation across a population into slices does not tell you what any individual can change. Second, the slices only add up neatly if genes, circumstances and activities do not interact or overlap, which is very unlikely. Third, no happiness questionnaire is perfectly reliable, and that imperfection alone means the neat 40% figure overstates how precisely anyone can pin down the room left for intentional activity.

Brown and Rohrer concluded that there is little empirical evidence for the pie's breakdown. To their credit, Sheldon and Lyubomirsky responded in 2021 by agreeing with many of the criticisms, while arguing that later research still supports their main point: people can raise their well-being through what they do, though the effects may be weaker than they first believed.

What this does and does not mean

It does not mean happiness is fixed. Genetic influence is real and substantial, especially on the long-term average of how you feel, but a heritable trait can still change, and population statistics cannot tell you your personal ceiling.

It does mean you should be wary of any precise percentage for how much of your happiness is under your control. Nobody has measured that, and the famous 40% is not a finding but a remainder. Most of the research also comes from Western countries, where environments differ in particular ways. If low mood persists for weeks, or gets in the way of daily life, a doctor or mental health professional is the right person to talk to.

What this means for you

A few grounded ideas follow from the evidence.

• Expect a baseline. Some people are simply sunnier than others on an ordinary Tuesday, and that is partly inherited. Comparing yourself to them is not a fair test.
• Do not wait for the pie. You do not need to know the exact size of your slice to try things that research links to well-being, such as time with people you care about.
• Look for interactions. Your temperament shapes which situations you seek out. Choosing environments that suit you is one way genes and choices work together.
• Be sceptical of neat numbers. A clean 50-10-40 split is a sign that something complicated has been simplified too far.

If you want a snapshot of where you are, our happiness test and wellbeing test measure different sides of feeling good, and the optimism test looks at how you expect the future to turn out, which is related but not the same thing.