1. Good ideas take time to bear fruit

In HLI, we advocate for taking happiness seriously: using the evidence on happiness to work out how best to increase it, then taking action. The case, in brief, is this: (1) happiness matters; (2) it can be measured simply by asking people how they feel; (3) research shows we’re often wrong about the degree to which things will impact our happiness. As far as we know, we’re the first and only organisation to look for the most cost-effective ways globally to improve happiness, and one of only a handful of organisations taking a ‘wellbeing approach’ to assessing impact. 

So, taking happiness seriously must be a new, radical idea, right? In 1809, Thomas Jefferson said, “The care of human life and happiness […] is the first & only legitimate object of good government”; in 1776, Jeremy Bentham said, “The greatest happiness of the greatest number is the foundation of morals and legislation”; and Aristotle said, over 2,000 years before that, “Happiness is the meaning and the purpose of life, the whole aim and end of human existence.” The idea may be radical, but it’s not new.

What is new is that, for the first time in human history, it’s become possible to pursue global happiness in a scientifically credible way – and so realise this Enlightenment dream. It was only in 1972 that the first large-scale surveys of happiness started in the US. Since then, the science of wellbeing has taken root and grown. There are hundreds of thousands of academic papers on happiness – in the social sciences, this is usually called ‘subjective wellbeing’. Many people know that the Nordic countries are supposedly the happiest on Earth, but the World Happiness Report, which produces the rankings, only started in 2012. Australia was the first government to collect data on its people’s wellbeing in 2002; now, over 70% of countries in the OECD, a group of mostly rich nations, do the same. Bentham and Franklin might have wanted to take happiness seriously, but they couldn’t. They didn’t have the data. Now, we do.

The Happier Lives Institute enters this historical arc in 2019. In the four years before, I’d been doing a PhD in philosophy at Oxford. I was inspired by the early Utilitarians, such as Jeremy Bentham and JS Mill, who had focused on the importance of happiness and, writing more recently, by Peter Singer, whose arguments had convinced me that those of us who can, ought to give some of our money to effective charities (even though I barely acted on this). But, while there was a growing demand for information on where effective givers should donate, something was missing. Where was the focus on happiness? Isn’t that what really matters? No one – including Singer, who later co-supervised my PhD – had tried to apply the empirical research on happiness to find the most efficient ways of turning money into happiness for other people. This seemed like an obvious gap, so filling it became a major task of my thesis. Among other things, I estimated that charities treating mental health in low-income countries (LICs) could be as impactful as, or even better than, the top charities beloved by those interested in effective giving – once we measured the impact on happiness. I realised far more and far better research was needed than I could do alone; thus, the Happier Lives Institute was born.

Five years later, we’re a team of nine spread across four continents. Our vision remains a world where everyone lives their happiest life, and our mission is still to guide decision-makers on how best to improve wellbeing. But, most importantly, we’ve made progress by assessing the effect and cost-effectiveness of a range of interventions on happiness for the very first time.

It took us until 2022 to compare the impact of four promising ways of helping people in LICs – treating depression, cash transfers, deworming, and anti-malarial bednets – in enough depth to make a recommendation to donors. We concluded treating depression was several times more impactful than cash transfers or deworming, so in 2022, we recommended StrongMinds, an NGO that treats depression at scale in Sub-Saharan Africa. We also concluded it was difficult to compare ‘life-saving’ organisations, such as the Against Malaria Foundation (which distributes insecticide-treated bednets), to ‘life-improving’ ones (like StrongMinds); this depends heavily on non-straightforward philosophical assumptions. As new information comes in, and our methodology improves, we update our results and our recommendations. In 2023, we reviewed our analyses, after which the cost-effectiveness of StrongMinds came down, so we recommended both StrongMinds and the Against Malaria Foundation; we also noted Friendship Bench, another mental health NGO, was a ‘promising’ organisation and flagged some other possibilities on the horizon. 

We’re delighted that donors have started to take note. We estimate that we influenced about $400k of donations in each of 2022 and 2023 (see Annual review 2022-3). At the end of 2023, a donor requested we set up a fund, Bloom, which will distribute $2m across 2024-5 to the most promising organisations (we’ll report on the grants for 2024 at the end of the year). 

This is a great start, and we’re excited about our trajectory. It’s taken time for ideas about the importance of happiness and effective giving to percolate through society; time for the field of wellbeing research to grow; and time for us to build on that research, share what we’ve found, and inspire people to act. The result is we’re collectively making progress on what we believe really matters: moving to a world where there is less suffering and more happiness. While the pace of change may feel slow to us, we don’t think it could have happened much faster and we’re confident the winds of history are behind us.

2. Finding the best ways to improve happiness is both and art and a science

HLI started with the aspiration to find the most impactful, evidence-based ways to improve happiness, but this was a fairly new approach. So, we weren’t sure how easy it would be or how well it would go. 

Here, in summary, is what we’ve done. We’ve taken academic research on randomised controlled trials (RCTs) of various poverty and health interventions in LICs, examined their effects on subjective wellbeing (SWB), aggregated the data, and created cost-effectiveness analyses (CEAs) to compare interventions and charities in terms of Wellbeing Life-Years, also known as ‘WELLBYs’. The UK Treasury approved the use of WELLBYs for policy appraisal in 2021; our approach is very similar. The great advantage of the WELLBY is that, by using a novel yet robust means of gauging the overall benefit of different interventions, we can discover unexpected, more effective ways to improve lives. As a result, we’ve identified some impactful, evidence-based ways to increase wellbeing so far and have shared our findings with donors.

In short, we’re all about RCTs, SWB, CEAs and WELLBYs. I hope you’re taking notes, there will be a test later…

Along the way, we’ve realised that CEAs, our bread and butter, are both an art and a science. Much of it is science: we draw on extensive data and then use various established methods from economics and psychology to draw conclusions. But there is an art to it, too: subjectivity is unavoidable, certainty is impossible, and philosophy continues to play a leading role. 

In our case, our analyses of spillovers (the effect an intervention has on people beyond the recipients, like, for example, others in a household) and the long-term effects of interventions are art and science; these are big factors, but data is scarce. Weighing the importance of different data sources is also a matter of judgement: which is more important, good monitoring and evaluation data from the charity itself or an RCT of a similar programme conducted on a different population in another country by a different organisation? There’s also the perennial question of ‘moral weights’, such as how to compare the quality and quantity of life. We conduct research into these thorny philosophical issues, too (e.g. this paper and this report). We aim to be explicit and thoughtful about our assumptions, but it’s important to recognise that reasonable people can and will have differing opinions.

The result is that while our CEAs are our best attempt to capture reality, we keep in mind that these are one perspective, and others attempting the same analysis might reach different conclusions.

3. Happiness research has made progress, but there is still far to go

HLI was an early adopter of the wellbeing approach to assessing impact. It’s been cheering to see how much research and interest in this field has accelerated in the last five years. Notable developments include the UK Treasury Greenbook recommending wellbeing for policy appraisal in 2021, New Zealand’s 2022 wellbeing budget, and Kier Starmer, who just became the Prime Minister of the UK, suggesting (in 2021) that policies be judged by their effects on both the economy and wellbeing. As I said in the opening, we should see these developments in the valuation and study of happiness as parts of a long-running story that began in the Enlightenment and will rumble on.

The field of wellbeing research has come far, but there’s still much further to go. We know quite a lot about and what’s good and bad for happiness. We have some understanding about how good or bad these things are; for instance, a doubling of income is associated with a 0.2 increase on a 0-10 scale and being diagnosed with depression is associated with a 0.7 drop. However, we know precious little about what the most impactful actions to take are, either for donors or governments. If we think of the problem like a jigsaw puzzle, it’s like a 1,000 piece puzzle, and we’ve maybe put together one corner and a few other pieces. 

While many countries measure wellbeing in their National Statistics Offices (NSO), no countries are yet taking wellbeing as the primary goal of policy. Clearly, there is a chicken-and-egg problem: if there isn’t a supply of research indicating the difference a wellbeing approach would make, why demand it? If there isn’t demand for it, who is going to supply the research? This underscores the need for continued research and advocacy to work on both the supply and demand sides.

4. The joys, sorrow and surprises of building a research non-profit

When people ask me how I’ve found the process of building HLI, I tell them not-really-jokingly that starting a non-profit has all the stresses of a normal startup but with no prospect of ever making me fabulously wealthy. 

This isn’t obvious to people, but non-profit and for-profit start-ups face very similar challenges. You have to think about your value proposition, product-market fit, budgets, pitching for funding, etc. It’s been hard work with regular challenges and crises of confidence. But working on the pursuit of global happiness is incredibly rewarding. And it’s all the better for having smart, kind colleagues who can help each other along the way.

Like any other start-up, we’ve had our share of growing pains and learning experiences. I’ll give some examples.

We thought our basic pitch about taking happiness seriously – we care about happiness, so let’s use evidence from happiness surveys and figure out what the priorities are – would be a much easier ‘sell’ than it turned out to be. Some people get it instantly. Others are far more suspicious. For the sceptics, it’s usually a combination of two things. The first is the sense that happiness, how good/bad people feel, isn’t deemed morally important; things like poverty and physical health are real problems, but unhappiness isn’t. The second is scepticism about whether we can put numbers on feelings, e.g. with a 0-10 scale.

To the first, we often find ourselves reminding – rather than convincing – people that happiness matters, both for themselves and others, and that poverty, physical health, and other problems matter not in a vacuum but because they affect happiness. To the second, we’ve lost count of how many times we’ve had to explain that subjective wellbeing surveys are not only scientifically validated but also, as Churchill might have said, that they are “the worst way of knowing how well someone’s life is going except for all the others that have been tried”. To sceptics, I often pose this question: if you wanted me to understand how your life is going, would you prefer I (1) looked at your bank balance and medical records, then guessed, or (2) asked, “How happy are you, 0-10”? It’s often only when people realise the alternative to (2) is using something like (1) that they see the value of the wellbeing approach.

We’ve also adjusted the tone and confidence with which we communicate our research. When we initially concluded that treating depression in the global south was possibly the most cost-effective way to improve happiness, which has been the result of a couple of years of hard research, we wanted to trumpet our success. We got pushback that we’d been overconfident in our conclusions, so we checked and double-checked our findings, and in the spirit of transparency, we’ve published anything we consider a ‘blunder’ on our suitably dubbed “Our Blunders” page.

One thing that’s going surprisingly well is running a remote team. We started doing this pre-COVID and found that the best people were spread out across the world. The whole team meets in person about once a year for a strategy retreat, and we have our all-hands call once a week where we share both work and personal updates.

5. Being hopeful for the future

I don’t know if he actually said it, but I heard that Bill Gates once said, in the context of international development, that we overestimate how much we can achieve in one year, and underestimate how much we can achieve in ten. We’ve now been conducting research and advocacy work for five years (longer in my case). Although progress often feels frustratingly slow, it doesn’t mean it isn’t happening or that you should give up. 

Over this time, we’ve seen many conversations go like this. It starts with scepticism about happiness – “What, happiness? That’s not a serious goal”. Then of measurement – “Fine, but you can’t really measure it, can you?” Then of the implications – “Okay, I guess you can measure it, but surely it doesn’t make a difference”. Then, to a grudging acceptance – “Well, I suppose it does make a difference, but maybe the difference isn’t that big.” And for some there is a final step, that of recognising the full implications – “Oh jeez, when we’re aiming directly at happiness, our priorities look pretty different. This is a big deal.”

The lesson here, for the advocates of seemingly weird ideas, is not necessarily to stop if you don’t get immediate traction. You should stop if the evidence stacks up against you, or perhaps pivot if something else seems more impactful, but merely running into opposition, or disinterest, isn’t sufficient to show that you’re going the wrong way. 

In our case, it’s pretty visible how attitudes towards wellbeing are on the march. The idea of prioritising wellbeing and making it a central societal goal is, as I said, at least 200 years old and stretches back to the Enlightenment. But, as Victor Hugo said, “Nothing is more powerful than an idea whose time has come.”