- 1. Why does subjective wellbeing matter?
- 2. Can we trust subjective measures of wellbeing?
- 3. Changing attitudes towards the importance of happiness
- 4. What’s wrong with the current measures?
- 5. Will our priorities change?
- 6. How can I contribute to a happier world?
- Further reading
If you want to do as much good as possible with the resources you have, here is one real-life choice you face.
For about $6,000 you could:
- expect to prevent one child from dying by providing 1,200 anti-malarial bednets;
- give a $1,000 cash transfer to six families living in extreme poverty; or
- treat 45 people for depression via group psychotherapy.
Which option will do the most good?
This is a challenging question – how can we possibly compare saving lives, reducing poverty, and treating mental illness? At the Happier Lives Institute, we believe it’s possible to answer this question by comparing the effects on people’s subjective wellbeing (self-reports of happiness and life satisfaction). What’s more, taking this approach leads to some surprising results.
The following sections introduce the key ideas and guiding principles that define our approach. Throughout, you’ll find many links to our growing list of research reports as well as the academic literature we draw on. Let’s begin with a general overview before expanding on each idea in greater depth.
We think that wellbeing is how people feel during or about their lives i.e. how happy or satisfied they are. When we talk about doing good, what most of us really have in mind is improving people’s wellbeing. But what is ‘wellbeing’? We believe that it is people’s subjective wellbeing that we should ultimately care about. Health and wealth aren’t important in themselves, they only matter because they improve our subjective wellbeing. Even if you think that happiness or life satisfaction are not the only things that matter, you will probably still care about them to some extent.
Subjective wellbeing can be measured in a scientifically valid and reliable way through self-reports. Social scientists have been doing large-scale population surveys since the 1960s. A typical question is “Overall, how satisfied are you with your life nowadays? (0-10)” Some people are very sceptical about putting numbers on feelings but we don’t think their objections are serious enough to give up on happiness surveys. If we want to know how people’s lives are going, we need to ask them.
Measuring and increasing happiness across society is a very old idea that was briefly abandoned and is now returning. Many ancient philosophers and Enlightenment thinkers understood happiness as the goal of life. But in the early 20th-century, the rise of behaviourism amongst economists and psychologists led to scepticism that we could or should measure feelings and we started using objective measures of social progress instead, such as GDP. In recent years, there has been a resurgence in support for a wellbeing approach amongst academics, policymakers, and civil society that is now gathering force. This seems to have been driven by several factors: mounting evidence that decades of economic growth have done little to improve average life satisfaction (a result known as the ‘Easterlin Paradox’), fears that economic growth is environmentally unsustainable, and greater confidence that measures of subjective wellbeing are valid and reliable.
It’s now possible – for the first time in human history – to start working out, in a scientifically rigorous way, the best ways to improve global happiness. Due to the rapidly increasing amount of wellbeing research conducted in recent decades, we now have a much better idea of the factors that have the biggest impact on subjective wellbeing. However, this work has barely begun and we don’t think anyone can confidently say what the top priorities are yet.
Current global priorities may change significantly if we adopt a subjective wellbeing approach. Of course, policymakers and philanthropists have already done a lot of thinking about how they can best use their resources to improve people’s lives. After all, haven’t we often made decisions with people’s happiness and life satisfaction in mind, even if we haven’t directly measured those things? Unfortunately, psychological research shows that we make predictable mistakes when thinking about what makes us, and others, happy. In broad terms, we place too much importance on what we can see objectively, such as wealth and physical health, and overlook people’s emotional and social lives.
We set up the Happier Lives Institute in 2019 to find the best ways to make lives happier. We currently focus on identifying high-impact funding opportunities in low-income countries because we expect that’s where money will go furthest. Our most thorough evaluation so far looks at the cost-effectiveness of providing psychotherapy for people living with depression compared to providing $1,000 cash transfers to very poor families. We estimate that psychotherapy is 9 times more cost-effective – the effect of therapy is slightly larger and it is much cheaper to provide. This challenges existing thinking about how to do the most good and suggests that treating mental health conditions should be a higher priority for decision-makers.
More research is needed to work out what the priorities should be. Mental health could turn out to be the most pressing global problem, but we may find other problems that are even more important. In the near future, we plan to evaluate a wide range of interventions including deworming programmes, cataract surgery, lead regulation, immigration policy, and access to pain relief. There are also some thorny philosophical issues related to the nature and measurement of wellbeing that need further thought, including how to compare saving lives to improving lives.
Our vision is a world where everyone lives their happiest life. We want to see a world where governments and philanthropists base their decisions, at least in part, on the best wellbeing evidence. If you think that making people happier sounds like a good idea, we encourage you to do what you can to make this happen. We provide a range of suggestions at the end of this article including advice for donors, researchers, entrepreneurs, and advocates.
1. Why does subjective wellbeing matter?
The claim that subjective wellbeing matters morally may seem to require no defence. But let’s put these ideas in their full theoretical context. In this section, we explain what wellbeing is, summarise the three philosophical theories of wellbeing, and introduce the three measures of subjective wellbeing that relate to each theory.
1.1 The three theories of wellbeing
In philosophy, wellbeing refers to what is intrinsically (or non-instrumentally) good for someone. Instrumental goods, like wealth, are only valuable as a means to something else, whereas wellbeing is what ultimately makes someone’s life go well. Understanding what makes life go well is of obvious value: every plausible ethical view holds that wellbeing matters in principle, and in practice, we do put great effort into improving the wellbeing of ourselves and others. Theories of wellbeing are generally divided into three families: hedonism, desire-fulfilment, and the objective list. The following sections provide a brief summary of each.
Learn more: read our explainer on the philosophy of wellbeing
Hedonists claim that wellbeing consists in an overall positive balance of pleasure over pain. In other words, what is good for you is whatever feels good to you. It seems implausible that something you haven’t experienced directly could make a difference to your wellbeing. If you never feel its impact, how could it affect your life?
The most well-known objection to hedonism is the ‘experience machine’ (Nozick, 1974). Imagine that you are offered the opportunity to spend the rest of your life inside a virtual reality capable of simulating any experience. Once plugged in, you could indulge in any number of pleasurable or happy experiences without ever wishing or even knowing you had left the real world. Hedonists think that plugging into the experience machine would make your life go better. But many people would reject the prospect of leaving behind the real world for a simulated existence, however pleasurable. This is often because they desire to do things in the ‘real’ world and the mere experience of (apparently) doing them is no substitute for in fact doing them. This leads us to the second family of theories.
Desire-fulfilment (or preference satisfaction) theories claim that it is the fulfilment of a person’s desires that makes their life go well. It is not the feeling or experience of a desire being satisfied that matters but that the desire is, in fact, satisfied. You want to actually climb Everest, rather than merely believe that you have. If you only wanted to feel satisfied, your view would also be vulnerable to the experience machine objection.
A general objection to this family of theories is that the objects of desire (friendships, success, happiness etc.) are not valuable because they are desired. Instead, they are desired because they are valuable, and so desire theories have got the explanation the wrong way around. Plant (2020) notes that desire-fulfilment theories are also vulnerable to ‘auto-maximisation’. This means that you can make your life better by simply choosing desires that are trivially easy to fulfil. For example, if your only desire is that Paris is the capital of France, hey presto, your life goes best. There also seems to be no way for these theories to apply to animals that lack desires about how their lives go overall, such as dogs, who we tend to think do have wellbeing.
Learn more: read our critique of desire-fulfilment theories
Watch: Michael Plant presents the critique to a seminar at the Wellbeing Research Centre
Objective list theories claim there are some things that improve your life that may be neither pleasurable nor desired by you. Classic items for this list include success, friendship, knowledge, virtuous behaviour, and health. Such items are ‘objective’ in the sense of being concerned with facts beyond both a person’s conscious experience and/or their desires.
Defenders of objective list theories might object to the previous two theories as naively simplistic – wellbeing cannot be reduced to a single element, life is far more complicated than that. Yet, pluralism, the view that more than one thing makes up wellbeing, faces its own challenges. If the items have a characteristic feature in common then you should replace the list with that single feature. If not, then how should you make trade-offs between the different items on the list?
1.2 The place of subjective wellbeing
Whichever theory you think is true, happiness and life satisfaction will play some role. For hedonists, happiness is all that matters. On desire theories, life satisfaction – how your life goes compared to how you desire it to go – is central and happiness matters because it is something we desire. If you favour the objective list, it would be odd if neither happiness nor life satisfaction were on it. Therefore, it’s not particularly controversial to claim that subjective wellbeing is something we care about. The more controversial claim, which we’ll come to later, is that we can measure psychological states through self-reports in a scientifically valid and reliable way
Even if you agree that subjective wellbeing matters, there are still questions about which theory of wellbeing is correct and determining the extent of practical disagreement between these theories is a further empirical challenge. The good news is that uncertainty about the correct theory of wellbeing does not prevent us from reaching some understanding of how to improve wellbeing in practice. This is because the three theories of wellbeing often agree on the things that lead to wellbeing: the person who is happy, successful, wise, and loved will have high wellbeing on all plausible theories. Nevertheless, the extent to which different theories of wellbeing suggest different priorities in practice remains an open question requiring further research.
1.3 Measuring subjective wellbeing
Subjective wellbeing (SWB) is an umbrella term for self-reported ratings of thoughts and feelings about life which fall into three main categories: hedonic, cognitive, and eudaimonic. SWB and happiness are often used interchangeably but SWB is a broader concept incorporating overall evaluations of life as well as momentary experiences of happiness. In the following sections, we describe each of these three categories in more detail and explain how they are measured. We leave discussion of why we think such measures are scientifically valid for Section 2.
All sensations can be placed on a single scale of ‘pleasantness’ and we define happiness as a net balance of pleasant over unpleasant experience. Using hedonic measures, the only two components of happiness are intensity (how pleasant/unpleasant something feels) and duration (how long the sensation lasts). Hedonic measures are sometimes split into positive affect and negative affect, which can each be measured separately and have different determinants. The former refers to pleasant emotions such as joy, contentment, and elation; the latter to unpleasant emotions such as sadness, fear, and anxiety.
Affect data is commonly collected using questionnaires based on PANAS and SPANE scales. Another measure is the day reconstruction method (DRM) in which participants break their previous day into episodes (like scenes in a movie) and state, for each episode, what they were doing, how they felt, and who they were with. A more accurate, but less common measure, is the experience sampling method (ESM) where participants are prompted, one or more times a day, to record how they are feeling at that particular moment. Although costly to implement, ESM avoids the problem of asking participants to remember how they felt. This is a significant benefit, given how error-prone our memories can be.
Cognitive (life satisfaction)
Cognitive measures of wellbeing are based on the view that happiness consists in having a favourable attitude towards one’s life as a whole. Life satisfaction is usually found by asking, “how satisfied are you with your life nowadays?” on a scale from 0 “not at all” to 10 “completely”. Most of the SWB literature has focussed on measures of life satisfaction up to now and there are two main reasons for this.
The first is practical. It is much easier to collect data on life evaluations than on experiences. Participants usually answer questions about life satisfaction in less than 30 seconds and it can easily be included in existing population surveys. By comparison, the ESM and DRM require more work from respondents; the former is intrusive and the latter takes respondents about 40 minutes to complete. Affect data from PANAS and SPANE questionnaires are strongly correlated with life satisfaction but the correlation is weaker for data collected using the ESM or DRM methods.
The second reason is moral. Life evaluations are sometimes thought by economists to be a measure of decision utility (what people choose to do). Economists have historically taken this to be of greater moral importance than experienced utility (how life is experienced).
Learn more: read our critique of life satisfaction measures
Watch: Michael Plant presents the critique to a seminar at the Wellbeing Research Centre
Eudaimonic (virtue and meaning)
Eudaimonic measures seek to quantify traits like virtue, character, and wisdom as well as concepts related to fulfilling our potential such as meaning, purpose, and flourishing. Eudaimonic measures are often regarded as a core component of SWB, particularly in the field of positive psychology. However, it is unclear whether measures of meaning are really measures of wellbeing and little data has been collected on them.
2. Can we trust subjective measures of wellbeing?
Perhaps you are sceptical of the claim that we can measure subjective concepts like happiness. However, the last few decades have seen an explosion of research on subjective wellbeing as social scientists and policymakers grow increasingly confident in the validity of these measures. This section provides a range of evidence that we hope will go some way to alleviating your concerns.
Before we begin, let’s introduce two key concepts: reliability and validity. A measure is reliable if it gives the same results each time (assuming the subject of measurement hasn’t changed). A measure that gave a random score each time would not be measuring anything. A measure is valid if it captures the underlying phenomenon it set out to capture. Let’s discuss each concept in turn.
Learn more: read our report on the measurement of wellbeing
A range of statistical tests has been conducted to confirm the reliability of subjective wellbeing measures. We quote here from the OECD Guidelines on Measuring Subjective Well-being (2013).
On hedonic measures:
“Diener et al. (2009) report…the positive, negative, and affective balance subscales of their Scale of Positive and Negative Experience (SPANE) have alphas of 0.84, 0.88, and 0.88 respectively. Krueger and Schkade (2008) report test-retest correlations of 0.5 and 0.7 for a range of different measures of affect over a two-week period.”
On cognitive measures:
“Test-retest results for a single-item life evaluation measure tend to yield correlations of between 0.5 and 0.7 for a time period of 1 day to 2 weeks (Krueger and Schkade, 2008). Michalos and Kahlke (2010) report that a single-item measure of life satisfaction had a correlation of 0.65 for a one-year period and 0.65 for a two-year period.”
Subjective wellbeing measures have also been subjected to a range of validation tests. The main test is construct validity – whether the measure performs in the world the way that theory predicts. Socialising, relaxing, and eating are associated with higher levels of positive affect, while commuting, working, and housework are associated with low levels of positive affect (Kahneman et al., 2009). Higher incomes are associated with higher life satisfaction and affect (up to a certain point) all around the world, at both the individual and country level (Stevenson & Wolfers, 2013). Stable, wealthy, well-governed countries score highly on average life satisfaction (Finland, Denmark, and Norway average around 7.5/10). On the other hand, war-torn and low-income countries do badly: South Sudan, Central African Republic, and Afghanistan average about 3/10 (World Happiness Report, 2021).
Hedonic and evaluative measures relate to different aspects of wellbeing so we shouldn’t expect them to give the same results all the time. For example, Deaton and Stone (2013) found that affect measures vary on the days of the week, improve with age, and only respond to income up to a point whereas evaluative measures correlate with income (even at high levels of income), are often U-shaped in relation to age, and do not vary over the days of the week. However, for practical purposes, affect and life satisfaction measures generally go together so we can use the latter as a substitute for the former. However, some caution is still needed as the measures differ in the degree to which different things matter and sometimes whether those things are positive or negative.
A long-standing worry about subjective wellbeing measures is whether the numbers represent the same thing to different people at different times (Ferrer-i-Carbonell & Frijters, 2004). For example, if two people say they are 5/10 happy, can we assume they are as happy as each other? More technically, the question is whether subjective scales are cardinally comparable – does a one-point change represent the same size change at all points of the scale? If the scales are merely ordinal (the numbers represent a ranking but contain no information on the relative magnitudes of differences), it will not be possible to use subjective scales to determine the best ways to increase global wellbeing.
The image below is taken from a 2018 survey by YouGov which asked people to score different words on a 0-10 scale from very negative to very positive. The results demonstrate that people tend to score positive and negative words in very similar ways which should increase our confidence in people’s ability to assign numbers to their subjective feelings in a meaningful and consistent way.
Learn more: read our report on the cardinality of subjective wellbeing scales
Watch: Michael Plant presents the paper at a workshop on the measurement of wellbeing
2.5 Other objections
Even if subjective wellbeing measures are generally good, it doesn’t mean they are flawless. But the fact there are some measurement issues is not sufficient to declare the scales invalid. If your bathroom scales give you an implausible result, you wouldn’t conclude that measuring weight is impossible. Much has been made of studies that find seemingly irrelevant factors can affect self-reports of wellbeing, such as finding a coin or being asked about your love life or politics just before you report your score. However, the impact of such factors is relatively limited and has not been borne out by the wider literature. Diener et al. (2013) find that 60-80% of the variability in life satisfaction is associated with long-term factors and the remainder with occasion-specific issues and measurement errors. When large populations are surveyed, random errors will wash out and such issues can also be minimised through careful survey design.
3. Changing attitudes towards the importance of happiness
Measuring and increasing happiness is not a new idea but efforts to conduct large population surveys only started a few decades ago. Why? What’s changed? In the following sections, we present a brief history of Western thinking about happiness over the last 2,500 years and highlight the current trends in wellbeing research.
3.1 The ancient philosophers
Aristotle used the term ‘eudaimonia’ for the ultimate goal of life. Wealth, health, and relationships were all considered to be important, but Aristotle placed particular emphasis on virtue as an essential component of a fulfilling life. Later schools of philosophy focused on specific aspects of Aristotle’s ideas. The Epicureans favoured a life based on friendship and simple pleasures whilst the Stoics emphasised the control of emotions in challenging situations. These two approaches to the good life spread throughout the Roman empire, aided by the Meditations of Marcus Aurelius, a Roman emperor and Stoic philosopher. However, as monotheistic religions spread across the world over the following centuries, people came to believe that happiness would be found in the afterlife as the reward for living a virtuous life on Earth.
3.2 The Age of Enlightenment
It was not until the late 1600s that philosophers began to re-establish happiness as the goal of life, with John Locke’s claim that, “the necessity of pursuing true happiness is the foundation of all liberty”. In the 18th century, Francis Hutcheson was the first to argue that we should produce “the greatest happiness for the greatest number”, an idea that influenced many great thinkers. Thomas Jefferson wrote that “the life and happiness of the people is the first and only object of good government”, Adam Smith studied economic methods to create the greatest happiness, and Jeremy Bentham coined the term utilitarianism in his famous book, The Principle of Morals and Legislation. John Stuart Mill inspired many social reforms based on utilitarian principles and the economist Francis Edgeworth discussed the need for a ‘hedonometer’ to measure happiness.
3.3 The sceptical turn
In the 20th century, the psychologists Ivan Pavlov and John B Watson claimed it was impossible to know what occurred inside people’s minds or how they felt. They argued that we could only study how people behaved and how this was affected by external stimuli. In 1932, Lionel Robbins (an economist) dismissed the idea that it was possible to compare the happiness of one person with another and argued that economists should only observe revealed behaviour, rather than attempting to measure how people feel. This turn to behaviourism led to GDP (gross domestic product) becoming the primary measure of national wellbeing in the post-war period, despite the protests of its inventor, Simon Kuznets, who said, “the welfare of a nation can scarcely be inferred from a measurement of national income”.
3.4 The return to wellbeing
Criticism of GDP as a measure of national progress has grown stronger in recent years. In 1974, Richard Easterlin showed that wellbeing in the USA had barely changed since WWII, despite significant economic growth (the so-called Easterlin Paradox). In a significant step forward for the field of wellbeing, Ed Diener and Daniel Kahneman (starting in the 1980s) showed that it was possible for wellbeing to be effectively measured and explained. Since then, psychologists and economists have made increasing use of subjective measures to investigate which experiences have the biggest effect on individual wellbeing.
The last few decades have seen an explosion of research on subjective wellbeing (SWB) with over 170,000 books and articles published in the last 15 years (see figure below). Policymakers are starting to take note too. The UK government has been measuring SWB for over a decade and, in 2013, the OECD suggested that its member-states should do the same. In 2019, New Zealand became the first country to introduce a Wellbeing Budget and joined Finland, Iceland, Scotland, and Wales to form the Wellbeing Economy Governments partnership.
Learn more: read Trends in conceptions of progress and well-being (World Happiness Report 2022)
4. What’s wrong with the current measures?
If subjective wellbeing measures have only come into use in recent decades, how have researchers and policymakers been measuring progress up to now? When a direct measure is unavailable, a ‘proxy measure’, one you think is a suitable stand-in, might be used instead. It is common to use income metrics (e.g. GDP per capita) or health metrics (e.g. quality-adjusted life years) as proxies for wellbeing but these suffer from a number of problems which we explain further in the following sections.
4.1 Affective forecasting
Psychologists use the phrase ‘failures of affective forecasting’ to describe the mistakes we make when predicting how we, and others, will feel in the future (Gilbert & Wilson, 2007; Wilson & Gilbert, 2005; Wilson & Gilbert, 2003). We underestimate the suffering caused by problems that are resistant to hedonic adaptation (Greene, Sturm, & Evelo, 2016), as well as experiences where it is difficult to put ourselves in the shoes of others (Igou, 2008). Our mental simulations produce these errors because they are egocentric, neglect context, and poorly represent time. Mental illness and chronic pain are two of the biggest problems that are under-valued in this way (Birkjær, Kaats, & Rubio, 2020; Graham, Higuera, & Lora, 2011). Understanding the “expectation-reality gap”, the gap between what people think makes others happy and what actually does, can help us to identify and evaluate global priorities more effectively. As you’ll see in the chart from Our World in Data below, people across a range of cultures mistakenly believe that others are less happy than they really are.
Learn more: read our report on affective forecasting
4.2 Moral weights
As we saw in the introduction, a challenging problem arises when you are faced with a choice between two (or more) different methods for increasing wellbeing. How much ‘moral weight’ should we give to saving a life, compared to increasing income, or treating someone for depression? Many people think it is too simplistic to reduce wellbeing to a single measure so they combine metrics for income, life expectancy, literacy, and many other things into a multi-dimensional index. However, each item on the list needs to be given an appropriate weighting which means that the difficult questions about trade-offs must still be faced. However, there is a solution to this problem. By measuring subjective wellbeing, we can directly compare the impact of different interventions on the wellbeing of the recipients rather than making an educated guess about their relative impact.
Learn more: read our report on estimating moral weights using subjective wellbeing
4.3 GDP and the Easterlin Paradox
GDP has been a common measure of national progress since the 1950s. However, the Easterlin Paradox (Easterlin, 1974) notes that average life satisfaction in the developed world has remained broadly stable over the last few decades despite a continuous rise in GDP per capita. This can be clearly seen in the charts below which plot life satisfaction (the white dots) against GDP per capita (the red line) in four high-income countries over recent decades.
This finding directly contradicts the evidence that richer countries are more satisfied than poorer countries and that richer people within countries are more satisfied than poorer people. The explanation lies in the fact that your wellbeing is not solely dependent on your own income, it is also affected by your relative income compared to your peers. For example, if you are wealthier than your peers then you should expect to have higher wellbeing. However, if your income rises, but the income of your peers also rises, these effects are cancelled out, leaving your wellbeing unchanged.
Coming soon: look out for our new post on the Easterlin Paradox
4.4 DALYs, QALYs, and WELLBYs
Economists often use health metrics such as the disability-adjusted life year (DALY) or the quality-adjusted life year (QALY) as a proxy for the quality of people’s lives. These are based on questions about time trade-offs, in which people are asked how many years of healthy life they would give up to avoid living with different health conditions. However, neither measure is an accurate guide to what makes people happy or satisfied because they are based on predictions of the badness of different health conditions rather than lived experience of those conditions.
If we only rely on guesses about future feelings, then we get the wrong answers about the effects of different health conditions on subjective wellbeing. For example, DALY estimates suggest that many people think that having depression would be about as bad as walking with a slight limp and they would trade off similar amounts of time to avoid either situation. However, data from people who actually live with these conditions shows that the effect of ‘moderate anxiety or depression’ is associted with a ten times greater change in subjective wellbeing than ‘some mobility issues’ (Dolan and Metcalfe, 2012).
Thankfully, there is a better solution. The ‘wellbeing-adjusted life year’ (WELLBY) allows us to compare different outcomes in a common currency (Frijters et al. 2020). One way, but not the only way, to specify the WELLBY is to say 1 WELLBY is equivalent to increasing life satisfaction for one person, by one point, for a year. This method is structurally similar to D/QALYs in that it seeks to combine quality and quantity of life. The main difference is that WELLBYs are based on direct reports of subjective experience, whereas D/QALYs are based on error-prone predictions of wellbeing in different scenarios. Also, note that D/QALYs are measured on a 0-1 scale whereas WELLBYs are based on a 0-10 scale.
Learn more: see how we used WELLBYs to estimate the moral weights of doubling consumption for one person for one year and averting the death of a child under 5 years old.
5. Will our priorities change?
As the previous sections have shown, research into the best ways to increase global wellbeing has barely begun. At HLI, we aim to find the most cost-effective, well-evidenced ways to improve subjective wellbeing and then communicate those to key decision-makers, such as philanthropists and policymakers, who can provide the resources to scale them up.
5.1 Our research so far
Our process for identifying high-impact funding opportunities is inspired by the effective altruism movement and the work of GiveWell in particular. As a result, our charity evaluation process starts with the expectation that donations will go furthest when helping those in extreme poverty, but we are confident of finding other funding opportunities that are even more impactful (see Section 5.2).
So far, we’ve evaluated two well-evidenced interventions: cash transfers and psychotherapy. A large number of academic studies show that cash transfers are a very effective way to reduce poverty and the strength of this evidence led GiveWell to recommend GiveDirectly as one of their top charities. However, we believed that providing group psychotherapy for people with depression in low-income countries could be even more cost-effective than cash transfers. After a rigorous search, we identified StrongMinds as one of the best charities delivering this intervention.
Drawing on evidence from over 80 studies and over 140,000 participants, we made a direct comparison of StrongMinds and Give Directly in terms of subjective wellbeing and found that StrongMinds is 9 times more cost-effective than GiveDirectly.
This is a surprising and important result. We’ve demonstrated that it’s possible to compare seemingly incomparable outcomes by measuring their effects in terms of subjective wellbeing. By doing so, we found a new and outstanding funding opportunity for philanthropists and policymakers. This provides further evidence that we shouldn’t rely on our intuitions about what makes people happier and should prioritise further research on the best ways to measure and increase global wellbeing.
5.2 More research is needed
We’re excited to explore other promising opportunities to improve subjective wellbeing on a larger scale. In this section, we set out the empirical and theoretical questions where further research could be most impactful.
Learn more: read more about our future research plans
We plan to examine interventions at three levels of scale:
- Micro-interventions (helping one person at a time) such as deworming, cataract surgery, digital mental health interventions, and cement flooring.
- Meso-interventions (object-level policies) such as regulation to reduce lead poisoning and air pollution, changing immigration policy in happier countries, and improving access to pain medication in low-income countries.
- Macro-interventions (large-scale societal changes) such as $20m+ funding opportunities and wellbeing policy ‘blueprints’ for governments.
In order to assess the cost-effectiveness of these interventions in terms of subjective wellbeing, we need data. Thankfully, there is now plenty of data available on the effect of certain interventions on the happiness, life satisfaction, and mental health of recipients. However, cost-effectiveness evaluations are sensitive to a number of other important factors for which the data is more limited and our modelling is more uncertain. These include the costs to deliver interventions, the effects of the interventions over time (which requires multiple, long-term follow-ups), the effects on other household members, and the household size.
Interventions provided to an individual often have ‘spillover’ benefits for other members of their household too. These spillover effects may even be greater than the direct benefit to the recipient. For example, in a household of four, if one member receives an intervention that improves their wellbeing it would only require a spillover effect of 33% for the household effect to be more important. If more data was collected, uncertainty in our cost-effectiveness analyses would be reduced and resource allocation would be more accurate.
Learn more: read our report on household spillovers for cash transfers and psychotherapy
Even with valid measures and sufficient data, identifying the best way to help others also involves some uncomfortable philosophical dilemmas and theoretical questions. One particularly pressing problem is defining the ‘neutral point’, the point where someone is neither satisfied nor dissatisfied.
If someone reports their life satisfaction score is 6/10 (on a 0-10 scale) is it better to: (a) extend their life for one year, or (b) improve their life satisfaction by one unit for two years? If the neutral point is 5/10, then (a) gives them one extra unit of wellbeing and (b) gives them two. However, if the neutral point is 0/10, (a) gives them six units and (b) gives them two. Hence, defining the neutral point is key to decisions about whether to extend or improve a life.
It is not clear where the neutral point is and there has been little discussion of how to determine this. Researchers sometimes treat the midpoint of subjective wellbeing scales (5/10) as the point where someone is neither satisfied nor dissatisfied, or neither happy nor unhappy. But if we take this as the neutral point, this has the controversial implication that many people, particularly those in low-income countries, have lives that are net negative (if we only consider their wellbeing). Other researchers treat the bottom of the scale (0/10) as the neutral point but this has a different, but also controversial, implication that it is impossible for anyone to have a life that is net negative.
Coming soon: look out for our working paper on defining the neutral point
Coming soon: look out for our philosophical review of Open Philanthropy’s Cause Prioritisation Framework
6. How can I contribute to a happier world?
There are many ways to increase the wellbeing of others through your career and/or donations. In this section, we highlight the most promising opportunities for impact.
There are many organisations doing a tremendous amount of good to improve wellbeing in the world. You can make an exceptional difference to the lives of others by donating thoughtfully to the most cost-effective and evidence-based charities. In addition to supporting our recommended charities, you can also fund further research by making a tax-deductible donation to the Happier Lives Institute.
One of the biggest challenges to increasing global wellbeing is establishing a strong evidence base of cost-effective interventions. Governments are understandably reticent to scale up unproven interventions, but promising interventions can only prove their value if they receive sufficient funding. Therefore, philanthropists and grantmakers have a vital role to play in supporting innovation and research.
Learn more: read our latest advice for donors
There are many potentially valuable research directions including: identifying the root causes of mental illnesses; improving preventive interventions; advancing the scale-up of existing treatments; and developing new treatments or improving existing ones e.g. psychedelic-assisted mental health treatments
If you’re a researcher that conducts studies to determine the effects of an intervention, please make sure you collect data on subjective wellbeing. And not just for the recipient; the effects on other members of the household can be just as significant. In order to conduct an accurate cost-effectiveness evaluation, it’s also important to measure the effects over time so plan to conduct follow-up surveys to determine how long the effects last.
Learn more: read our research agenda for the most pressing questions that need to be answered
Non-profit organisations serve a very valuable role, particularly in low-income settings where governance is weak, such as conflict-affected regions (where the prevalence of mental disorders is exceptionally high) or in locations where the local government has no reach, such as rural or remote areas. NGOs can also be innovative in ways that governments are less likely to be, for example by finding new ways to improve effectiveness or lower costs and may be able to encourage governments to scale-up services. For example, Sangath has worked with state governments in India to train community health workers to deliver low-cost mental health interventions at scale.
Given the size of the treatment gap and the scope for innovation, we’re excited to see new non-profits being set up to experiment with different delivery methods. Charity Entrepreneurship, a non-profit incubator, provides training and guidance to those wishing to start new charities. In 2020, they conducted a detailed search for interventions that improve mental health and subjective welľbeing.
Tech entrepreneurship also seems very promising. Examples include Headspace (a meditation app), Happify (happiness-training exercises), MindEase (anxiety relief), UpLift (CBT for depression), and Sanvello (a support platform for mental health). Digital mental health is still in its infancy. However, if effective and commercially viable products can be delivered at scale, it would make an enormous dent in the burden of global mental illness.
Advocates can raise awareness of neglected global priorities like mental health and press policymakers to provide more and better services. This could include campaigning, journalism, lobbying, or work in think tanks at a national or international level. Even those engaged in research and intervention delivery can undertake advocacy work alongside their other activities.
A key role that advocates can play is bringing together diverse actors and stakeholders. Different groups, such as clinicians, civil society, and economists, take different approaches to advocacy which can result in them passing contradictory messages to decision-makers. Improvements in advocacy could involve greater coordination of these messages, clearer communication to decision-makers, and better use of evidence of the efficacy of psychosocial interventions in low-income countries to inform decision-makers.
There are tremendous opportunities for policymakers that wish to increase the wellbeing of their populations as cost-effectively as possible. Where funding and political will exists, they can scale up the evidence-based interventions developed and tested by researchers and non-profits. And even if no additional funding is available, there are likely to be many ways to use existing budgets more effectively, such as shifting mental health expenditure from psychiatric hospitals to community-based services.
Thank you for exploring our key ideas. We hope you learned something new and feel inspired to make a change to your work and/or donations based on our research. If you’d like to explore the field of wellbeing science in more detail, you’ll find some of the most important books, reports, and journal articles in our reading list.