# 86

This (cross-)post is an update from the Happier Lives Institute (HLI) and contains our 2020 Annual Review. In it, we describe and explain the research we conducted in 2020; comment on our research evaluation procedure, the outcomes of our work (that we are aware of), and some lessons learned; lay out our research priorities and plans for 2021, and our longer-term vision; describe ways you can support us on our mission.

This review is about 5,000 words in length. We have provided an executive summary and note some readers may want to check the contents sidebar, then go straight to specific sections, rather than read the whole thing in order.

# Executive Summary

HLI searches for the best ways to measure and increase global well-being; we believe that subjective well-being (SWB) scores (self-rated happiness and life satisfaction) are both a highly promising and under-utilised tool for this purpose. 2020 was HLI's first full year of research and operations as an organisation - it was an opportunity to focus on addressing the most important questions on our research agenda, and to develop our capacities as an organisation that conducts careful and action-relevant research.

The ‘backbone’ of our work is assessing the value of different outcomes in terms of  ‘Wellbeing-Adjusted Life-Years’ or ‘WELLBYs’, an approach recently pioneered in economics that we are applying and developing. In our post on ‘moral weights’ we explain the method and use it to compare, for the first time, the relative values of two actions that are particularly relevant for the effective altruism community: doubling household consumption for a year and averting the death of an under-5, both in low-income contexts.

We worked on the theoretical underpinnings of the WELLBY framework by producing: a working paper on the nature of well-being; another on the comparability of subjective scales; and a report on the validity of SWB (to be published soon).

We also applied the WELLBY methodology to evaluating interventions and programmes. We conducted a systematic review and meta-analysis on the impact of cash transfers on SWB, and started work on other interventions - cataract surgery, lay-delivered psychotherapy and positive education. Further, we reviewed two cause areas - pain and mental health (coming soon) - as a way to scan the horizon for potentially highly cost-effective interventions. We also continued to work on our mental health programme evaluation project, aiming to identify impactful donation opportunities in global mental health.

We mainly saw 2020 as a year to establish and develop our research capacity, but we are pleased that various academics and organisations in the effective altruism community are already using our work. Founders Pledge, for example, used the results from our ‘moral weights’ post in their internal prioritisation of charities.

Our main focus for 2021 is to continue applying the WELLBY framework to various ‘micro’-interventions. We will continue to investigate whether using SWB indicates new priorities for the effective altruism community. Further, we hope to further demonstrate that WELLBYs enable comparisons to be made between a wide range of interventions, and search for new, potentially highly cost-effective interventions.

As we demonstrated in our moral weights post last year, this analysis is highly sensitive to various moral assumptions, such as the views taken about what well-being consists in, the badness of death, or population ethics. We plan to present the ‘results’ for various viewpoints, so readers can make their own moral judgements.

We also plan to continue our foundational work on measuring well-being, for example, by investigating how to compare existence to non-existence using subjective well-being scales (estimating the ‘neutral point’).

Finally, given HLI has (implicitly) focused on helping humans in the ‘near-term’, and the increasing interest in longtermism, we plan to do some exploratory work developing our views on longtermism and what that would imply for the value and direction of our project.

Our five-year vision is to further establish ourselves as a research institute that acts as a bridge between SWB researchers, the effective altruism community, and other institutional decision-makers, producing clear, useful, and rigorous research that leads to happier lives around the world.

We are currently recruiting for a Senior Researcher or Researcher, a Head of Operations and Communications, and a small cohort of Summer Fellows. Find out more

Finally, we have room for more funding over the next year and are looking to expand our donor base.

# Introduction

The Happier Lives Institute (HLI) searches for the best ways to measure and improve global well-being; we conduct global priorities research using the lens of subjective well-being.

HLI was founded in mid-2019 and went through the Charity Entrepreneurship incubation programme that summer, growing from 1.5 to 2.5 full-time equivalent staff. We then fundraised and hired a full-time researcher at the end of the year.

2020 was our first full year of research and operations as an organisation. It was an opportunity to focus on addressing (what we considered to be) the most important questions on our research agenda, and to develop our capacities as an organisation that conducts careful and action-relevant research. Our aims were (1) to build the case for using subjective well-being as an outcome measure in impact evaluation and (2) to investigate whether using this new approach would indicate new priorities for the effective altruism community. During this time our 2.5 FTE research staff were supported by a team  of dedicated volunteers across the year.

The review is divided into three main sections:

1. A discussion of the research we conducted in 2020
2. A reflection on the impact of our research so far
3. A summary of our future plans

# Our outputs in 2020

The projects we worked on in 2020 are described below, in an order that we hope provides clarity on how they relate to one another. For each project, we briefly describe what we did, what we learnt, and who found it useful (as far as we know). For all projects, please refer to the articles themselves for further information; we intend this post to offer high-level reflections rather than summaries of the results.

## Estimating the value of different outcomes using subjective well-being (the WELLBY framework)

It is common in impact evaluation to measure changes in outcomes like health or income, and use these as indicators of quality of life. But really we care about how people’s lives are going for them overall. Well-being-adjusted life years, or ‘WELLBYs’, provide a ‘common currency’ with which to compare different outcomes against each other, such as averting deaths, alleviating poverty, enhancing education, and improving mental health. This is a similar idea to the well-established quality and disability adjusted life year (QALYs and DALYs) health metrics, which combine quality and quantity of health into a single number.

In 2020, GiveWell informed us that they were reconsidering their approach for comparing the value of outcomes, what they call ‘moral weights’, and were interested in our approach. Two of GiveWell’s moral weights in their cost-effectiveness analysis are the relative values of (1) averting the death of an under-5-year-old and (2) doubling consumption for one person for one year. Therefore, we felt this was an excellent opportunity to present the WELLBY framework and how it can be used to estimate these moral weights, as well as make an initial estimate of the weights and see how it changed the results. Read the post.

The moral weights are highly sensitive to some of the underlying philosophical assumptions, particularly the views of the badness of death and population ethics. We decided to present the analysis for different viewpoints, allowing the reader to make up their own mind about which view to take; we felt this was both transparent and flexible. We covered ‘deprivationism’ and the ‘Time-Relative Interest Account’ (TRIA) as the accounts of the badness of death under a person-affecting view of population ethics, since we thought these were most likely to match GiveWell’s stance. In future work, we plan to expand these viewpoints to include a totalist view of population ethics and an Epicurean view of the badness of death. See sections 1.3.2 and 1.3.3 in our updated research agenda for further information about these views.

Finally, we chose to use life satisfaction as the ‘base unit’ of the WELLBY, because much more life satisfaction data are available than happiness data (‘life satisfaction’ refers to how you judge your life to be going overall, ‘happiness’ to how good/bad you feel during it). In our cash transfer meta-analysis (a later project, see below), we found more happiness data than we were expecting, so we hope to be able to assess outcomes using (at least) happiness and life satisfaction data in future. Whether one thinks happiness or life satisfaction is what ultimately matters may alter the priorities; this was a further topic we investigated (see below).

In recent conversations with GiveWell and Founders Pledge, we learnt that Founders Pledge used our result to compare across charities for their internal purposes of prioritisation, and that GiveWell is considering to what extent to use SWB to determine their moral weights (see GiveWell’s conversation notes, which implicitly indicate where our perspectives may differ).

In many ways, the WELLBY framework forms the backbone of our work, and our other projects are about either improving the quality of the analysis or extending it.

## Evaluating interventions and programmes, and ‘horizon scanning’

The WELLBY framework demonstrates how, in principle, to use subjective well-being in impact evaluation. We also want to put this into practice and provide insights into the most cost-effective ways to increase well-being.

We decided that ‘micro’-interventions in low- and middle-income countries were a sensible place to start, for several reasons:

• A significant amount of money (around $150 million/year) is moved to GiveWell recommended charities; it would be impactful if our work helped to improve this allocation or to identify new giving opportunities. • Micro-interventions are simpler to evaluate than policies or other ‘macro’-interventions (aka ‘systemic changes’), which means we can focus on the methodological questions around using SWB specifically. • Existing work, particularly by GiveWell, has already made substantial progress on evaluating the cost-per-outcome (e.g. the cost to GiveDirectly to provide a cash transfer), which means we can concentrate on how to use SWB to estimate the value-per-outcome (i.e. the moral weight of a cash transfer). It is likely that we will expand the kinds of interventions we study and/or the settings of those interventions in future. You can read more about our five-year vision below. The first intervention we studied was cash transfers, which are a helpful baseline for other interventions. Furthermore, the SWB literature on the link between income and SWB has long lacked much causal evidence (it has mostly been correlational), which means randomised studies looking at cash transfers are a way to answer the age-old question of whether money makes people happy. We decided to conduct a systematic review and meta-analysis, i.e., searching all the available literature on the effect of cash transfers on subjective well-being in LMICs, and submit it to an academic journal. We recognised this would be considerably more work than a non-systematic, non-academic review, but we felt that the expertise we would gain from working with our two social science collaborators at the University of Oxford, as well as the academic peer-review process, would be beneficial to HLI. Additionally, we recognised that the work would be particularly useful to the wider academic community in this case, given the large number of studies (38) that hadn’t been synthesised, and to policy-makers in developing countries and the development sector. For a summary of our findings, see this page. In late 2020, we also started working on other interventions - cataract surgery, and psychotherapy delivered by non-specialists - and continued work on ‘positive education’ (providing psycho-social education in schools). Our plan for each intervention is to write up a review of the evidence of the effect of the intervention on SWB and to construct a cost-effectiveness model. In addition to our work on interventions, we have been searching for the most cost-effective mental health programmes that individuals could donate to (read more). We plan to publish a charity evaluation later this year. We will also publish a standalone review of the evaluation project. One point we want to make now is that we learnt that (1) predicting the marginal cost-effectiveness of a donation to an organisation implementing a programme, is a much more involved task than (2) estimating the retrospective cost-effectiveness of an intervention described in the academic literature. Having said that, donation recommendations are clearly more actionable to donors than intervention reports. We expect intervention evaluations to be useful to other institutions that make recommendations, such as GiveWell, Founders Pledge, and Charity Entrepreneurship, and charity evaluations would act as guides for individual donors as well as informing institutions; the extent to which we focus on interventions rather than organisations remains to be seen. Finally, we worked on two ‘problem area reports’ on pain and mental health in 2020 (the former has been published, the latter will be published very soon). These projects serve as ways of ‘horizon scanning’ - taking a broad look at a cause area with an eye out for potentially cost-effective programmes or policies. By communicating our findings, we also hope to push forward the understanding of individuals and organisations who may be interested in working in the area. The pain report highlighted advocacy and policy interventions that are examples of the kind of interventions we may evaluate in future. ## Theoretical work on the nature and measurement of subjective well-being Using SWB in impact evaluation relies on some assumptions about measuring well-being. We produced working papers on two questions in this category in 2020, and wrote a report on a third question. We felt these were important internally and they had been raised as important questions in a survey of key stakeholders in effective altruism that we conducted at the end of 2019. The first question was whether subjective scales (e.g. rating happiness 0-10) are cardinally comparable: does a one-point change, on a given scale, represent the same size change for different people at different times? Among key stakeholders, this was the major objection in principle to using SWB. They were more-or-less prepared to accept that ‘happiness surveys’ did measure happiness, but they doubted that individuals’ numerical scores could be meaningfully aggregated in the way necessary for cost-effectiveness analysis. It turns out this question had received little attention in academia either, perhaps because it's quite theoretical for social science, and quite empirical for philosophy. In fact, it had received so little attention it wasn’t clear how to tell if there was a problem or not. A summary of the working paper is available here. As a result of this work, we had a small, positive update on the cardinality of current SWB data and a moderate-large positive update on the problem being solvable in principle. We were able to break down the problem into its various ‘moving parts’, i.e. the key assumptions, and say how to empirically assess them. Our empirical review found evidence supporting cardinality and no strong evidence against it. We sketched some ideas for how further work could ‘correct’ for issues, if they are later found. We are pleased that, we believe in part due to our work, academics seem to be taking more note of this question (see the series of seminars from the Oxford Wellbeing Research Centre on “fundamental issues in the measurement of SWB” on 4 February 2021, in which we participated). Charity Entrepreneurship has also commented that they found the paper useful for their work using SWB and other subjective scales more generally. We are currently collaborating with several academics to write a more technical paper on this topic to submit to an economics journal. The second question was on the nature of well-being: what is it that constitutes our ‘well-being’, that is, what is ultimately good for us, that makes our lives go well? Among SWB researchers the dominant view seems to be that life satisfaction (a judgment of how life is going overall), rather than happiness (how good/bad life feels), is what ultimately matters; within the effective altruism community, the reverse seems true - happiness is considered to be the priority - although several people we’ve spoken to seem torn. We wrote a paper investigating the nature and plausibility of life satisfaction theories of well-being, available here. The main practical conclusion was that life satisfaction theories face some underappreciated or unrecognised objections and, as a result, are likely implausible. To be clear, life satisfaction scores can be (and, we think are) a very useful indicator of well-being, even if well-being doesn’t consist in life satisfaction. Our inclination is to think happiness is what ultimately matters and life satisfaction scores are useful, in practice, as information about that, particularly given how much more life satisfaction data there are. We hope these arguments spur SWB researchers to pay more attention to measuring happiness. As SWB research is a young field, we see a narrow and urgent window of opportunity to influence its underlying assumptions before they get bedded in for years to come. We intend to maintain some level of academic work; in general, we suspect that HLI will have the most impact acting as a bridge between academia, effective altruism, and other stakeholders (see our five-year vision, below). In addition, we worked on a review of the validity of SWB, that is, the degree to which SWB metrics measure what they purport to measure. This was also mentioned as being important in our survey of stakeholders, who were sometimes sceptical it was possible, even in theory, to measure happiness (or other subjective states). The report traces the development of validity theory in the social sciences to its current form, construct validation, where a measure is deemed valid if it behaves as expected, given scientists’ best understanding of the underlying construction. The construct validation seems reasonable to us and, given large swathes of (social) science rely on it, not easy to abandon. We then conducted our own, non-original review of the validity of experienced measures of SWB (i.e. happiness, rather than life satisfaction) and judged them to be sufficiently valid, by the standards of social science. The report will be published in the near-future. Finally, we also wrote a short ‘explainer’ for a general audience that outlines the philosophical accounts of well-being - hedonism, desire satisfaction and objective list theories - and explains some of the arguments for and against each of these theories. ## Other activities # Evaluation of our year ## External reviews As a research organisation, it is challenging to track and evaluate our impact. Nevertheless, we think it is important to try. We have been using three methods so far: 1. Eliciting reviews of our main research outputs from experts We ask subject-matter experts for candid reviews of the quality of our research. Last year, we received reviews of our research from Rossa O'Keeffe-O'Donovan (Global Priorities Institute, University of Oxford), Julian Jamison (University of Exeter), and Aidan Goth (Founders Pledge). The reviews were favourable overall and provided useful suggestions for improvements; you can read them here, here and here. We also received journal peer-reviews for the academic papers (which we are not able to share yet). 2. Conversations with key stakeholders It is important to make sure our research is not just rigorous, but actually useful. We have been speaking to key stakeholders with the aim of understanding their own priorities and how our work could be useful to them. This is also an opportunity to find out whether our work has changed their decision-making in any way. Tracking these ‘wins’ is an important part of how we evaluate our own impact; the most important outcomes that we know of are given in the next section. 3. Conducting an impact survey This is an opportunity for our followers and members of the EA community to give us feedback on: the quality of our work; if and how they updated their beliefs or changed their actions based on it; and indicate their preferences for our future work. At the end of 2020, 31 people filled in the survey. Because this is a small sample, and many respondents are already likely to be HLI supporters, we are not putting much weight on the results, but we were pleased with the scores provided on the quality of our work and whether respondents had updated positively towards using SWB in impact evaluation. We want to continue to develop our evaluation processes. We will aim to collect a larger and wider sample of responses in all three areas. We might experiment with methods of ‘open peer review’, such as sharing reviews publicly (which would hold ourselves more accountable) and having ways for the wider community to contribute (to encourage more candid reviews). ## Our impact As mentioned, we mainly saw 2020 as a year to establish and develop our research capacity, rather than make confident recommendations for individuals and organisations. We are pleased that, nevertheless, various academics and organisations in the effective altruism community are using our research: • GiveWell are considering to what extent to use SWB to determine their moral weights, which we believe is - at least in part - motivated by our work. • Founders Pledge used our ‘moral weights’ result in their internal charity prioritisation, and use subjective well-being in some of their charity evaluations. • Charity Entrepreneurship used life satisfaction as a main outcome metric in their work. They also investigated mental health charities based on our reasoning that it might be unduly neglected, and Canopie - an organisation providing mental health programmes for mothers - was founded as a result. • 80,000 Hours referred to our work on its problem profiles and list of important research questions. • The Vox ‘Future Perfect’ newsletter (of December 4th) shared our cash transfer working paper. • We believe our work has helped to motivate the recent mini-conference hosted by the Oxford Wellbeing Research Centre on “fundamental issues in the measurement of SWB”. Whilst these outcomes do not immediately translate into happier lives, we take them as some external indication of our progress. If you know of anyone else who has used our work, please let us know by emailing clare@happierlivesinstitute.org ## Other lessons learned We identified the following growth areas in a recent internal evaluation of 2020: • Projects suffered from planning fallacy and scope creep. In general, writing up a project was the stage that we most egregiously underestimated how long it would take. Scope creep was more of a problem when we were vague on the audience for a given report. • In general, we aim to be clearer internally on the intended audiences for each research project, and to involve relevant stakeholders earlier in the project, to ensure that our work is as useful as possible. • As our timelines extended, mid- and long-term planning got dropped. We recognised this in mid-2020, however, and introduced quarterly OKRs, which improved the situation. • We probably should have spent more time on fundraising, and developing our fundraising strategy. Our biggest bottleneck at the moment is our empirical research capacity. We expect we could have fundraised and hired more staff, rather than having our current staff do as much direct research themselves. # Plans for 2021 ## Research A major project of HLI is estimating the value of outcomes from various interventions in terms of WELLBYs, and then informing decision-makers. This is a major empirical enterprise: doing cost-effectiveness analysis in terms of SWB is basically new, so we're often having to scrape existing studies to create these estimates ourselves (as we did last year with cash transfers). However, it's not just an empirical exercise. There are foundational and not-yet-well-investigated issues in philosophy of science about how and whether to measure subjective states that we need to get the bottom of (e.g. cardinality). And this analysis is, unsurprisingly, potentially sensitive to various moral questions, such as, which measure of well-being should be used? We plan to inform decision-makers when and where the priorities are altered by moral judgments rather than facts. A further task is to provide information on how to choose in the face of those decisions. In our updated research agenda, we divide this work into three projects; area 2 is our main focus and where the majority of our effort will go. To bolster our applied research capacity, we are currently advertising for a Researcher or Senior Researcher; we will also run a cohort of summer research fellowships. More information on these is given below. Area 1: Foundational research into measuring well-being • Examining how best to convert between different SWB, as well as other, measures. • Investigating how to compare existence to non-existence using subjective well-being scales: determining the ‘neutral-point’. Area 2: Applied research to identify and evaluate the most cost-effective ways to increase well-being • Estimating, in terms of SWB, the impact of potentially highly-effective interventions, including: psychotherapy for common mental disorders; cataract surgery for blindness; and deworming tablets to improve lifelong earnings. • Setting out how different moral assumptions—about what well-being is, the badness of death, and population ethics—alter those cost-effectiveness estimates and may alter the priorities. We think evaluating the cost-effectiveness of interventions in terms of SWB has a high value of information for the following reasons: • By investigating how sensitive priorities are to the outcome metric used, to explore what difference using SWB makes. • Enabling a comparison of cost-effectiveness estimates for interventions that had previously been hard to compare in a non-arbitrary manner (for example, cataract surgery vs cash transfers). • Indicating where we might find new, highly cost-effective interventions. Area 3: Understanding the wider global priorities context Although HLI “conducts and promotes research into how best to measure and increase global well-being”, in practice, our focus is on helping humans in the ‘near-term’. We note there is a growing interest in longtermism, a perspective we have not engaged with so far. We plan to do some exploratory work developing our views on longtermism and considering what it would imply for the value and direction of our project. For a full description of our future research projects, and our motivation for them, please read our updated research agenda ## Summer fellowship HLI started as a volunteer group and has benefited enormously from individuals offering their time. Until 2020, individuals mostly approached us ad hoc with their project proposals. Since then, we moved more to a model of recruiting specifically to further our research priorities. In 2021, we plan to run a summer fellowship, and recruit individuals to work on short (6-10 weeks FTE) research projects, and be compensated for their time. We will write several project proposals as well as invite applicants to make their own suggestions of projects that will contribute to our updated research agenda. See information below about how to apply. ## Become registered as a UK charity We are applying to the Charities Commission to be a UK charity. Our board will be: • Peter Brietbart (Chair): Peter is the CEO of the anxiety management app Mind Ease. He is a veteran of the effective altruism community with a background in philosophy, entrepreneurship and project management. • Samantha Bernecker: Sam is a Clinical Research Scientist at Happify Health. She earned her PhD in Clinical Psychology from the University of Massachusetts Amherst in 2017 and completed a postdoctoral fellowship at Harvard University. Before joining the team at Happify Health, Sam worked as a management consultant at Boston Consulting Group. • Barry Grimes: ​Barry is an Events Associate at the Centre for Effective Altruism, which produces the EA Global and EAGx conference series. He has over 10 years experience working in communication and stakeholder engagement roles across the public, private and not-for-profit sectors. # Developing our five-year vision We see our work as global priorities research using the lens of subjective well-being. Whilst our current applied work is fairly narrow - focussing on micro-interventions in LMICs - the potential applications of our research are wide-ranging. We can imagine expanding our analysis to systemic interventions, policies, or considering implications for the long-run future. We would like to further establish ourselves as a think-tank that acts as a bridge between SWB researchers, the effective altruism community, and institutional decision-makers. We plan to explore other networks, such as the international development sector or public policy, to strategically expand our reach. We believe that a better understanding of how to measure and increase well-being is not only achievable, but can be successfully communicated to decision-makers, both public and private, who will be motivated to act. The result should be a substantial, wide-ranging, and long-lasting improvement in the experience of life, one that will not be achieved without deliberate effort. # How you can help Stay informed Sign up to our newsletter for quarterly updates on our work and to hear about future vacancies. You can also follow our work on Facebook, Twitter, and LinkedIn. Make connections If you, or someone you know, has expertise in the areas on our research agenda then we would love to speak to you, or them! Please reach out at michael@happierlivesinstitute.org. Provide feedback If you have any questions or comments about our work, please share them in the comments section below or contact us directly at michael@happierlivesinstitute.org. We’re excited to answer your questions and learn from your experience. Come and work with us Apply to join our team! We have several opportunities: • Senior Researcher or Researcher • If you’d like to work on our applied research agenda, apply before 23rd May • Head of Operations and Communications • If you’re a smart generalist who wants to enable and streamline our work, and ensure it’s communicated effectively, apply before 23rd May • Summer Research Fellows • If you’re interested in working on questions on our research agenda for 6-10 weeks this summer, then please fill in this form to express your interest. We will email you when we post project adverts (we welcome project proposals as well) and open applications. Provide funding Reasons to fund HLI: • We have made good first steps in establishing ourselves as an organisation doing rigorous research on important questions. • We believe the research questions we are going to address over the next year have a high value of information and that we have demonstrated promising progress on these so far. • There are encouraging signs that organisations will utilise our work, leading to better decision-making and happier lives. • We have been supported by Charity Entrepreneurship, the Effective Altruism Infrastructure Fund (as well as when it was the EA Meta Fund), an anonymous six-figure donation (thank you!), as well as a number of generous individuals in the effective altruism community. In total, we have raised £295,000 in funding to date, and have filled our budget until the end of 2021. We have a projected funding gap of £150,000 until the end of 2022, and do not yet have known donors to fill it. The following descriptions illustrate what we could do with different sums: • A donation of £8,000 would cover the costs of three summer fellowships. An example of a research project would be an investigation and review into when happiness and life satisfaction data suggest different priorities. • A donation of £50,000 would pay for a Senior Researcher for at least one year, who could search for high-impact programmes, widening the pool of donation opportunities and increasing our ability to give action-oriented advice to a number of institutions. • A donation of £100,000 would cover the costs of the salaries of the three current staff members for one year, ensuring our work safely continues and our 2022 priorities are achieved. We are hoping to find one or two large ‘strategic’ donors, as well as broaden our base of small to medium donors, so that we can continue our work and plan for the future with confidence. If you are interested in donating, please make a donation or email michael@happierlivesinstitute.org. # Appreciation Michael, Clare and Joel would like to thank everyone who has supported HLI on its journey so far. In particular, the volunteers who have contributed to our 2020 research: Kaleem Ahmid, Jide Alaga, Lucia Coulter, Derek Foster, Ulf Johansson, Milo King, Florian Kuhlmeier, Carmen Csilla Medina, Sid Sharma, Jasper Synowski, Suzanne Van Arsdale, Sonia Vallentin, and Caitlin Walker. A huge thanks to the board: Peter Brietbart, Sam Bernecker, and Barry Grimes. We are also very grateful to the Charity Entrepreneurship team for ongoing support and mentorship, and to Julian Jamison, Rossa O’Keeffe-O’Donovan, Aidan Goth, John Halstead, Matthew Walker, Tina Khanna, Johannes Haushofer, and several others for reviews or feedback on our work, and to Caspar Kaiser and Anders Malthe Bach-Mortenson for collaboration. Finally, huge thanks to all of our funders. Thanks for reading and have a happy day! # 86 Pingbacks 10 comments, sorted by Highlighting new comments since New Comment It's cool to see a new strand of priorities research being developed! (I also appreciated the clarity and readability of the update.) Thanks Ben, we're glad you enjoyed the post! Good to hear what you're up to, seems like a very promising project to me! A few questions about WELLBY though 1) How do you envision WELLBY interacting with the SWB community? Is the long term plan that this will remain a metric developed by you or is the long term plan for this to be a more independent measurement that for SWB researchers use and improve upon without any involvement from you 2) Do you see it being a long term "competitor" to other metrics such as QALYs and DALYs as they are used in global health today, or do you imagine it being something that only organizations in EA or perhaps mental health use in the foreseeable future? 3) how does "Well-being-adjusted life years" get shortened into "WELLBY" rather than "WALY", "WBALY" or at least "WELLBALY"??? :( Did you consider the negative effects on the reader's WELLBY caused by the confusion about how that abbreviation works? :'( Hello Engelhardt, Thanks for the comment! In response to your comments: 1. To clarify, the WELLBY is something that has come out of the academic SWB community - bits of economics and psychology, mostly. It's not been developed by us, as there are only a handful of papers that have used it so far; hence we're among the first to be applying it. I should add that, if you're already using measures of SWB, say, a 0-10 life satisfaction scale, it's not a big innovation to look at how much something changes that, then multiplying that by duration, which is really all the WELLBY is. (The more innovative bit is using SWB at all, rather than using WELLBYs given you're already using SWB.) So, it easiest to think of us as using a relatively new, but existing, methodology and applying it to new problems - namely, (re)assessing the cost-effectiveness of things EAs already focus on. That said, there are some theoretical and practical kinks to be worked out in using WELLBYs - e.g. on the ‘neutral point’, mentioned above. Our plan - which we are already engaged in - is to do the work we think is necessary to improve the WELLBY approach, then feed that back into SWB academia. More generally, it’s not unusual that a measurement tool gets developed and then refined. 1. Ideally, we’d like to see SWB metrics used across the board, where feasible, and we are pushing to make this happen. Part of the issue with Q/DALYs is that they are measures of health. Even if you thought they were the ideal measures of health (or, the contribution of health to well-being) you run into an issue comparing health to non-health outcomes. A chief virtue of SWB metrics is that you can measure changes in any domain in one currency, namely their impact on SWB. Having said this, Q/DALYs are quite ingrained in the medical world and it’s an open question how valuable it is to push for change their vs do other things. 1. I think the rules can be bent in search of a good name, and we're really just following what other SWB researchers call them. It has been suggested, notably by John Broome, that it should be the 'WALY', but that sounds a bit, well, silly (in British English, a ‘wally’ is a synonym for ‘fool’). Personally, I also like the SWELLBY, but that’s yet to catch on... Congratulations on an excellent year of research and thank you for your contributions to this cause area! To offer a perhaps under-explored research thread in the treatment and mitigation of cluster headaches and migraines, which I haven't seen in pieces from HLI or OPIS, there appears to be moderately high efficacy (for n=18) of a ketogenic diet on reduction in incidence of such headaches: "Of the 18 CCH patients, 15 were considered responders to the diet (11 experienced a full resolution of headache, and 4 had a headache reduction of at least 50% in terms of mean monthly number of attacks during the diet)" (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5816269/). Keto diets are quite restrictive, require an adaptation period (anywhere from 2 days - 2 weeks, the above study utilized an 18-day period, https://hvmn.com/blogs/blog/ketosis-how-long-does-it-take-to-get-into-ketosis-and-keto-adapt) and are often not particularly animal wellfare-friendly. I'd be interested to explore the cost-effectiveness of an intervention based on using exogenous ketones, especially those derived from plants (somewhere around$1-2 per day? e.g. https://shop.perfectketo.com/products/powdered-mct-oil-from-coconuts-for-keto-and-ketosis, https://rspnutrition.com/products/ketobhb) as an alternative to drug-based treatment of CH. Although research is sparse, it seems probable that exogenous ketones also decrease the adaptation period to a keto diet (ethically important due to the extreme nature of these conditions). There are some reported, rather mild side effects to exogenous ketones (https://perfectketo.com/why-ketones-cause-stomach-pain/), but these seem negligible in the case of treating CH and migraines.

Thanks for your comment and for bringing this to our attention. One of the pleasures, but also pains, of SWB research, is that there is simply an enormous scope of it; basically everything impacts well-being one way or another. The result is that many potentially fruitful avenues of research are left unturned.

I don't expect we'll be pursuing this specific line of inquiry, or headaches in general, with the next year or so. The only scenarios in which I would see that change would be if (1) a major donor appeared and would (only) fund us to look at headaches or (2) we already had a lot of donors following our recommendations - we don't have any such donors now, which is necessarily the case because we don't have any all-things-considered recommendations(!) - and our inside view what the headaches might be more effective than our hypothetical top pick and so worth investigating.

As a hot take on your particular suggestion, this is a very small study and I've heard lots of horror stories about dietary research, so this causes me only a (very) minor update, sorry!

Love to hear that there is important work being done in that area! Are there approaches to measure SWB as a function of "objective" well-being (OWB)? And what are their shortcomings? For instance, to me it feels that SWB could be a weighted sum of OWB the recent change in OWB and how one's own (x) OWB compares to the OWB of others (x_i):

The weights are probably person specific parameters. Persons with low w1 and w2 might be resilient, persons with high w2 might like status symbols,  etc.

I'm not sure exactly what you mean by "objective well-being". Here are two options.

One thing you might have in mind is that well-being is constituted by something subjective, eg happiness or life satisfaction, but you then wonder how objective life circumstances (health, wealth, relationship status, etc), positional concerns, etc. contribute to that subjective thing. In this case, health, etc are determinants of well-being, not actually well-being itself. This approach is pretty much exactly what the SWB literature does: you see how the right-hand side variables, many of which are objective,  relate to the left-hand side subjective one. I'm not sure what the shortcomings of this approach are in general - if you think well-being is subjective, this is just the sort of analysis you would want to undertake.

An alternative thing you might mean is that well-being is properly constituted (at least in part) by something objective. One might adopt an objective list theory of well-being:

All objective list theories claim there can be things which make a person’s life go better which are neither pleasurable to nor desired by them. 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

If one had this view, your question would be about how well-being, which is objective, relates to how people feel about their well-being. It's not clear what the purpose of this project would be: if you already know what well-being is, and you think it's something objective, why would you care how having well-being causes people to feel about their lives? So, I assume you mean the former!

On a similar note, how does HLI relate to Seligman's well-being concept and PERMA measurement? In Flourish, he argues: "Well-being theory denies that the topic of positive psychology is a real thing; rather the topic is a construct - well-being - which in turn has several measurable elements [PERMA], each a real thing, each contributing to well-being, but none defining well-being (...) By contrast, life satisfaction operationalizes [="defines exhaustively"] happiness in authentic happiness theory [measured by life satisfaction]"

I am aware that life-satisfaction is currently a more realistic thing to incorporate as a measurement into intervention and that PERMA is covered in the Positive Education paper.
Thank you very much in advance!

And thank you for the transparency regarding HLI's work both when it comes to research and organisational strategy. I found them highly useful for my own work and aspirations.

Finally, a slightly random question: Which part of Layard's book did HLI contribute to? I just finished it a month ago. :)

Hello!

I'm not really sure what Seligman means in the above quote, sorry. Perhaps it would make sense in a wider context.

Re PERMA, I'm not a fan of the concept and it strikes me as unmotivated. It's something like a subjective list theory of well-being, where Seligman takes well-being to consist in a bunch of different items, each of them subjective in some way. However, I don't see the justification for why he's chosen those 5 items (positive emotions, engagement, relationships, meaning, accomplishments) rather than any others. It seems to be the most plausible re-interpretation of PERMA is that those 5 items are major contributions to happiness, and well-being consists only in happiness.

I'm glad you like our transparency! We hope it helps us improve our decision-making and better allows others to see how we think.

Re Layard's book, Richard asked me to read a draft and I gave him extensive comments, primarily on the philosophical aspects, which were mostly in the earlier chapters. I also attended a conference he put on to discuss the book.