What should a billionaire do to maximise human happiness?

Reading time: 11-14 minutes

[Note to reader. This is a draft 5,000-word essay that will eventually form a chapter in my thesis. I’ve cut corners on some important issues to save space and I’ve had to signpost that I would deal with them elsewhere. I hope this is still useful in its current format. I wrote this 6 months ago and whilst I’ve changed my views since, particularly on the value of saving lives (see this) I haven’t incorporated such changes yet.]

In this paper I consider what a billionaire committed to classical utilitarianism would do to maximise the happiness of currently existing humans. I see this as an effort to develop the debates surrounding effective altruism, a philosophy and social movement which encourages people to do the most good they can with their lives.[1],[2] At present, effective altruism largely focuses on three areas: eradicating poverty and physical illness in the developing world (hereafter I’ll simply call this ‘poverty’), increasing animal welfare and reducing existential risks to humanity. Respectively, these can be explained in terms of a concern for present human life, present sentient life, and future human life.[3]

The main objective of this paper is to argue effective altruism has so far overlooked an alternative option of trying to increase the happiness of present people: by changing how they think. I call these sort of programmes ‘internal happiness interventions’ (‘IHIs’) in contrast to the ‘external happiness interventions’ (‘EHIs’), such as poverty reduction, that aim to increase happiness by improving the environmental conditions of individuals’ lives. I argue the hypothetical billionaire should fund experimental IHIs which target mental illness and what I call ‘ordinary human unhappiness’, the sort constituted by everyday stress, sadness, misery, etc., as these are very likely to be a more cost-effective way of increasing happiness than any EHIs effective altruism presently advocates.

The structure of this paper is as follows. In section 1 I specify my terms and explain why effective altruists who are concerned with the suffering of living humans focus on poverty reduction. In section 2 I focus on the unhappiness effective altruism has so far ignored – mental health and ordinary human unhappiness – and present empirical evidence suggesting they are very substantial causes of unhappiness. In section 3 I discuss what sort of IHIs could be tried and why there’s a strong prima facie case to expect them to be more cost-effective than any present EHIs. In section 4 I highlight a set of objections I don’t have space to discuss here, but would expect to resolve in the thesis itself. I conclude in section 5.

  1. Classical utilitarianism, effective altruism and what effective altruists recommend currently

There isn’t a canonical effective altruist position, so I’ve decided to approach the question of how to maximise the happiness of present humans from the perspective of a billionaire committed to classical utilitarianism. Doing this allows us to get a clear sense of what the happiest world would look like which non-utilitarian positions are then able to deviate from in as much as they value other things. I’m focusing on the happiness of presently existing humans only as I don’t want to get side-tracked into the comparative value of animal life to human life, or the value of the present to the far future. I’ve specified billion pounds because that allows me to consider large programmes a small donor would not countenance, but I’m not going to try to provide a detailed allocation of resources.

Classical utilitarianism is understood as having three components. [4] First, hedonism about well-being: what makes someone’s life go well(/badly) is experiencing happiness(/unhappiness) – as opposed to having one’s desires met or achieving items on an objective list – and every moment of experience has the same importance to their well-being. I’ll define ‘happiness’ here as any mental state that feels good to the person feeling it, and unhappiness as the converse.[5] Second, the additive method of aggregation: the best state is affairs is the one where the sum of the individuals’ well-being is the largest (as opposed to best state depending on how well-being is distributed). Third, maximising consequentialism: that we ought always to bring about the best state of affairs (as opposed to an improved, or satisfactory, state of affairs). In short, we ought always to maximise happiness.

Effective altruism, however, lacks a precise formulation. We can define thick and thin versions of it.[6] The thin version holds that you should ‘do the most good you can’ by using one’s spare resources to help others without espousing a particular account of the good or one’s relation to it. The thick version is quite similar, but not the same, as utilitarianism. It largely subscribes to maximising consequentialism and the additive method of aggregative, but adopts a welfarist, rather than just hedonist, theory of value, such that met desires and objective facts can have value. The thick version also holds decisions should be guided by the best available evidence, such as randomised controlled trials, to understand what works.

For the sake of simplicity, I’ll assume that effective altruists are classical utilitarians in disguise. I’ll drop this assumption only at the end when I consider how an effective altruist with some prioritarian or egalitarian sympathies might respond to my argument.

Effective altruists argue the best way to do good is to identify large problems and then find cost-effective ways to tackle them.[7] In practice, they typically conclude this means donating to anti-malaria and anti-poverty charities, such as the Against Malaria Foundation and Give Directly, respectively.[8] Poverty and malaria are both very big – 702 million people live on less than $1.90 a day[9] and 146 million suffer from malaria each year, of which over 400,000 die[10] – and very bad: when we think of those in extreme poverty, or convulsing with malarial paroxysms, it’s not hard to imagine their suffering.[11],[12] Such people are typically neglected – their own governments are poor and most charity raised in the developed world is spent domestically – and they can be helped at minimal cost: £3 ($5.60) will buy a malarial bednet and $1.90 a day brings someone out of poverty.[13] Hence they look like the biggest causes of suffering to currently existing humans but also, because the people they affect are so poor, offer very cost-effective ways to increase happiness by providing some basic goods.

  1. The unhappiness effective altruism ignores

As far as I can tell, effective altruism has not thought seriously about focusing on mental ill-health and ordinary human unhappiness as causes, even to discard them.[14] I’ll now explain what I mean by the two terms and outline why they are significant.

Mental ill-health is probably familiar to the reader, but the terminology around it can be confusing. I’m referring to what are sometimes called ‘common mental disorders’ – depression, bi-polar, anxiety, schizophrenia, bulimia, anorexia, and attention hyperactive defective disorders.[15] Given that mental illnesses comprise many negative mental states, they are clearly quite bad for happiness.

By ‘ordinary human unhappiness’ I mean the sort of everyday emotional suffering we experience as a result of fear, worry, sadness and so on. I also take it to refer to the idea that simply because someone is neither unhappy, nor diagnosed with a mental health disorder, that does not mean they are as happy as they could be. Typically, we don’t consider the sort of frustration we experience from heartbreak or tedious commuting worthy of moral attention, perhaps because we don’t think we can do anything about it, but it’s relevant on a utilitarian framework. (I note the distinction between mental illness and ordinary human unhappiness is arbitrary and nothing hangs on its precision: mental illnesses and ordinary human unhappiness are supposed to highlight different points on the happiness spectrum. I could alternatively have called these something like ‘clinical unhappiness’ and ‘non-clinical unhappiness’ instead.)

The number of people affected by these is certainly large. 753.4 million people a year suffer from mental ill-health, which is slightly less than the 858 million affected by poverty and malaria combined (on the false assumed there is no overlap between poverty and malaria). Mental illnesses, such as depression, are obviously very bad for happiness.[16] Ordinary unhappiness is a problem all 7.125 billion existing humans suffer from, and whilst such emotional suffering may be not acute (if it were, it would be classed as mental illness) it is frequent and persistent.

That effective altruism has overlooked these seems indicative of the moment’s focus on external happiness interventions. To put this in context, if we want to increase happiness, it seems there are only three options: changing how people think, their objective circumstances – their wealth, health, physical environment, etc. – or how they spend their time. I label these internal, external and temporal happiness interventions respectively. (‘IHIs’, ‘EHIs’, ‘THIs’). To give some concrete examples, I could try to make you happier by teaching you to be more grateful so you appreciate your life more (an IHI), encouraging you to spend more time on things you enjoy, such as socialising, and less of things you don’t, such as working (an THI) or I could move you to a bigger house, hoping you’ll enjoy time in your new environment more than your old one (an EHI). As happiness is a mental state, all happiness interventions are successful if and only if they cause mental state the person would have experienced to be replaced with a more positive one, either by re-evaluating the original experience more positively or re-allocating attention to a more positive experience. Therefore, although, on one level, all happiness interventions change individual’s thoughts, we can distinguish IHIs, EHIs and THIs by what they target: for instance, if I try to get you to see your friends more I am trying to target how you spend your time (a THI), without changing any of the external facts of your life (an EHI), or how you consciously evaluate your thoughts or allocate mental attention (an IHI).

To be clear, the anti-poverty and anti-malaria charities effective altruism currently recommends are EHIs because they try to improve the objective circumstances of individual’s lives – their wealth and their future health, respectively – in order to make them happier. This might have an impact on how they think or spend their time, but those are not what is targeted. My argument is that effective altruism might find it more effective to instead focus on IHIs as they directly change how people think. IHIs would focus on mental illness and it’s less acute, non-clinical sibling, ordinary human unhappiness. I’m not going to discuss the THIs in this essay – more research is required to see how effective they could be – but I also note that effective altruism has not countenanced the use of any THIs either.

The obvious question that follows this analysis is whether IHIs are a more cost-effective method of increasing happiness that effective altruism’s current EHIs. At present, I don’t think we know. I think it’s very possible that they could be, which is why I argue the billionaire conducts a set of trials to find out. To understand why IHIs are comparatively promising, we need to turn to the empirical research on happiness. Presuming this essay forms a chapter in my thesis, I would have already discussed the research on happiness and its limitations in an earlier chapter(s) and I would refer to it again now. I will only provide an abridged form of the evidence here.

Research into happiness is typically (but not always) described as measuring ‘subjective well-being’.[17] Subjective well-being is normally understood as having an evaluative and an experiential component.[18] For the evaluative component, individuals are typically asked ‘on a scale of 1 to 10, how satisfied are you with your life overall?’ which is usually called a ‘life satisfaction’ measure. For the experience component, individuals might be asked to report on their emotional experiences, often from yesterday, which is sometimes called a measure of ‘positive and negative affect’ (‘affect’ is understood as ‘emotional state’). I would have argued that the latter is closer to the hedonic conception of happiness I would be interested in in virtue of adopting a classical utilitarian outlook. In practice, far less research has been conducted using experience (i.e. happiness) rather than evaluation (i.e. life satisfaction) measures, so we are forced to use the latter as a proxy for the former, despite the fact they produce somewhat different results.[19] I would identify how they differ and also argue, despite the misgivings philosophers may have, that happiness does appear to be usefully measurable.

There are two general findings about happiness we need to take account of for practical ethics, which psychologists term ‘hedonic adaptation’ and the ‘failures of affective forecasting’.[20],[21] The former refers to the fact humans are very good at adapting to change, such that most events only cause a temporary change in happiness, rendering most EHIs unpromising. The latter refers to our inability to correctly imagine how happy we will be in the future, or other people are in other circumstances: whilst we tend to be correct on whether something will make us happy or sad, we display an ‘impact bias’, overpredicting the duration and intensity of events.

Adaptation seems sufficiently strong it’s questionable if giving money to even very poor people will make them happier. So far only one randomised control trial has been conducted into the topic: a study into the impact of Give Directly’s cash transfers to Kenya farmers showed recipients’ life satisfaction levels returned to pre-donation levels after 6 months, implying the donations had no impact over the long-term in terms of life satisfaction (experiences of happiness were not measured).[22] Whilst this is only one study, the results should not be understated: Give Directly’s cash transfers appear to have no long-term impact on happiness at all. Speculatively, it might be the case that Give Directly’s recipients were in secure poverty – they simply don’t have many possessions – and not in insecure poverty – fearing for their survival – and only taking people out of insecure poverty increases happiness.[23] Nevertheless, Give Directly’s current programmes seem not to be raising happiness. This result fits the other findings on adaptation. One example is the so-called Easterlin Paradox, which finds life satisfaction scores have not increased, or have barely increased, as the world has become richer.[24] Richer people are shown to be more satisfied with lives, but not much happier, than poorer people, suggesting wealth has relative, rather than absolute, effects on happiness and life satisfaction, and people’s standards rise in step with development.[25] Some doubts have been raised about this finding, which I would have had to discuss. Studies also find if people become disabled they experience a major short-term drop in life satisfaction but adapt partially to almost totally after a few months.[26],[27]

I would argue these results are surprising because of our difficulties in making successful affective forecasts. Largely this seems to be a focusing effect or ‘focusing illusion’: when we think about the poor, we focus on their poverty, failing to realise that’s not what they spend their time thinking about, and that we would eventually adapt to their circumstances as they have.[28] I would have explored both hedonic adaptation and affective forecasting in greater depth earlier in the thesis to get a sense of their functioning and limitation. For instance, people don’t seem to get used to the negative experience of longer commutes, so moving house and then having a longer commute would therefore be EHI that brought about non-adapting unhappiness.[29]

This analysis of happiness also suggests our current measures of health states are skewed. At present, effective altruists rely on Quality Adjusted Life Years (QALYs), a metric which combines the quality and quantity of life that healthcare interventions generate. QALYs express different health states on a scale between 0 (death) and 1 (full health) which can then be multiplied by how long those states last for. In the UK the National Institute of Clinical Excellence (NICE) recommends the “value of changes in patient’s health related quality of life should be based on public preferences using a choice-based method”[30] which are calculated by asking the public for their time trade-offs (TTO) preferences: how many years they would trade-off to be cured of some health state.[31] On this analysis, studies find people would give up as many years of remaining life, about 15%, to be cured of some ‘some difficulty walking’ as they would to be cured of ‘moderate anxiety or depression’.[32]

The flaw with this methodology is that not only is it hard imagine mental illnesses without experiencing them – it’s easier to visualise malarial fever than depression – but people fail to account for the fact they will adapt to some conditions but not others. For those interested in happiness, a more useful way of measuring ill-health is using subjective well-being measures (i.e. life satisfaction and daily affect).[33] Using this method, people are asked about their subjective well-being and their health states separately, so they aren’t focusing their attention directly on their health conditions. In stark contrast, using subjective well-being scores, ‘moderate anxiety or depression’ is associated with 10 times a greater loss to life satisfaction, and 18 times a greater loss to daily affect, than ‘some difficulty walking’ is.[34]  This implies moderate anxiety or depression is about 18 worse for happiness than we imagine it to be. The subjective well-being approach shows severe anxiety or depression three times worse than severe pain when measured in terms of daily affect (the closest proxy for happiness) which contrasts with the result that people expect severe pain to be worse than severe depression or anxiety.[35] I’ve included a table displaying these results in Annex A.

Both hedonic adaptation and the failures of affective forecasting suggest EHIs will be less useful that we intuitively thought. When we imagine the poor, we focus on their poverty from our perspective, forgetting they will have (largely) adapted. When we imagine malarial paroxysms, we forget that this is not how someone who catches malaria spends their average day, most cases of malaria are similar to influenza, and 99.5% of people survive malaria.[36] It’s questionable if the billionaire should spend any money on poverty, at least until he sees further evidence (which he should fund the collection of) and malaria, whilst bad, is likely less bad than he would pre-evidentially have thought. In contrast, it doesn’t seem that we could adapt to depression – clearly, experiencing unhappiness will always feel bad – or that we are good at understanding emotional suffering we cannot see. Further, the small, frequent misery we experience through ordinary human unhappiness might be quite substantial over a lifetime. Hence IHIs, which would target mental health and ordinary human unhappiness, already look relatively a lot more promising. By taking some guesses at the relative severity of the different problems and the numbers of people involved, I estimate mental illness and ordinary human unhappiness causes between 4 and 72 times more unhappiness annually than poverty and malaria combined. I’ve attached my calculations in Appendix B, but note these figures are not should not be taken literally, they are only to get a rough sense of scale.

  1. What effective altruists could and should do instead

The next question the billionaire will want to ask is how cost-effective IHIs are comparted to EHIs. If mental health cause 10 times more unhappiness than malaria, but mental health is not as cost-effective to treat, the billionaire would do more good if he spent his money on malaria.

I’m now going to argue there are a range of promising potential IHIs the billionaire could trial that may be cost-effective. Regarding mental health a number of methods which have been shown to work including, but not limited to, Cognitive Behavioural Therapy (‘CBT’),[37] mindfulness-based stress reduction (‘MBSR’)[38] and, to a much lesser extent, Positive Psychotherapy.[39] The basic ideas behind each of them is that CBT teaches people to understand their thoughts and stop negative thinking patterns, MBSR can be understood as a form of meditation that helps people accept, rather than fight, negative emotions and so reduce the suffering they cause, and Positive Psychotherapy trains people to find more positive emotions, such as by encouraging people to be grateful, forgiving, or engaging in things that give them fulfilment. All three of these methods have also been shown to work on mentally ‘healthy’ (i.e. ‘non-clinical’) populations too, and would therefore help reduce ordinary human unhappiness.[40],[41],[42] Somewhat conveniently it looks like the same set of internal happiness interventions could be used for the whole population and would help mentally ill and mental healthy people.

Whilst all these methods have previously been delivered in-person by therapists, they could also be provided digitally, via computer or smartphone, where individuals teach themselves how to be happier, which could be a lot cheaper. There is evidence electronic therapies work too, albeit they seem to be less successful for more severe problems.[43] It would also be hard to combine digital methods with the provision of psychiatric medication.

As such, if the billionaire wants to maximise happiness, there are a number of options available for him to try. The three most obvious are:

Digital Campaigns: public happiness campaigns are trialled in developed and developing countries. These make people aware of, and pointed towards, digital versions of CBT, MBSR and/or positive psychology. These would be seen by both those with and without mental illnesses and so would help people with both high and low levels of happiness.

Mental Health Charities: working with existing charities, or starting new ones, to test the most effective ways of providing on the ground, in-person mental health services in developing countries, possibly including the use of cheap, off-patent anti-depressants.  This would primarily help those with the lowest happiness.

Lobbying: lobbying governments of developed countries to do all of the above, where relevant, in addition to pushing for emotional resilience training to be included in schools and more funds for conventional mental health treatments.

There may be other options. The question remains as to whether these would allow the billionaire to do more good than funding anti-malaria programmes. There’s lots of uncertainty on this, but it looks like current treatments for malaria are (according to Givewell) about $100/DALY[44] whereas treatments for depression are $1,000-$3,000/DALY[45] in the developing world. DALYs – ‘Disability Adjusted Life Years’ – can be understood as the reverse of QALYs: QALYs measure years in perfect health gained, DALYs measure years in perfect health lost. I argued such measures may underrate the happiness lost to mental health by, very approximately, 10-18 times, which means funding Mental Health Charities may already be the most cost-effective way of increasing happiness.

However, I think the billionaire should be optimistic he could do much more good than he would by funding anti-malaria programmes. Mental health in the developing world is very neglected and so there is likely very large scope for innovation. One third of Lower and Middle Income (i.e. developing) Countries[46] do not have a designated mental health budget,[47] and for those that do the average expenditure is 0.5% of their total health budget,[48] and the treatment gap (the number who don’t get treatment as a percentage of those who need it) is 76-85%.[49] A Centre for Global Development report describes it as “truly neglected area of global health policy”.[50] At present, there is only one large charity, Basic Needs, working in developing world mental health. I conclude that there might be substantial scope for Mental Health Charities to innovate and provide care more cost-effectively than they do already.

Digital Campaigns may be more effective than Mental Health Charities. Without requiring health professionals to administer treatment, Digital Campaigns could reach huge numbers of people and scale easily. My guess is that the Digital Campaigns may be more successful in developed countries as there is less stigma and more awareness around happiness and mental health, as well as better access to technology. I’m uncertain about estimating the cost-effectiveness for Digital Campaigns or Lobbying, so I won’t venture an answer. However, given that the Bill and Melinda Gates Foundation, a huge philanthropic organisation, funds lobbying as well as in-country programmes that implies serious philanthropists conclude lobbying is a cost-effective way of supporting their work.[51]

So what should our utilitarian billionaire do if he wants to maximise world happiness? It looks like the most promising options would be to run a number trial IHIs. He should fund Digital Campaigns in developed and developing world and Mental Health Charities in the developing world to find out how cost-effectively they can increase happiness. These should be randomised controlled trials using, amongst other things, the measures of happiness I mentioned earlier (I would have investigated measuring happiness at greater length in an earlier stage of the thesis). He should also run randomised controlled trials to assess how much happiness is increased by anti-poverty and anti-malarial interventions – it still seems somewhat implausible anti-poverty programmes cannot make people happier. Given the scale of mental health and ordinary human unhappiness (see Annex B) and the uncertainty surrounding how effective it could be treated, the initial cost of funding trials seems justified. If he’s spending money on anti-poverty and anti-malaria programmes he should stop, wait until he has more evidence of where to do the most good, then decide where to allocate the rest of his resources. (Effective altruism also seems to suffer from the ‘billionaire problem’– if a charity like the Against Malaria Foundation doesn’t reach its fundraising target a billionaire will step in to fill it such that, counterfactually, your donation may do no good – thus our imaginary billionaire withdrawing his funds is unlikely to cause the suffering one would predict.[52])

It’s worth stating the obvious point of why he is committed to this position. As a classical utilitarian his only goal is to maximise happiness (in this case, of humans) without regard for how that happiness is distributed. If he was previously funding anti-poverty or anti-malaria programmes he should move his funds to test new IHIs because they may allow him to do more good. If the billionaire finds he can bring about more happiness with a Digital Happiness programme in the developed world rather than a Mental Health Charity in the developing world, he should do so. My conclusion is that effective altruism, understood as classical utilitarianism in disguise, is currently not currently pursuing the best opportunity to do good. It has overlooked Internal Happiness Interventions (as well as Temporal Happiness Interventions), and this is what a classical utilitarian billionaire should now trial.

  1. Objections

There are a number of potential objections to my conclusion that the billionaire should switch his focus EHIs from IHIs. I don’t have space to discuss them in any depth here, but I list them below and will have to address them in the thesis itself.

  • Summing of small gains

The thought here is that it’s morally objectionable to fund digital happiness programmes which cause a small gain to many instead of anti-poverty or anti-poverty programmes which save lives. The classic statement of such a view is Scanlon’s, who argued it would be wrong to give someone painful electric shocks to allow thousands of others to watch a football game (this is slightly different, as Scanlon is not comparing small gains with lives saved, but small gains with large losses, and also making a point about the wrongness of gladiatorial sports as entertainment).[53]

I think such a criticism would be a misunderstanding of my position. Digital Campaigns would reach both mentally ill and mentally healthy people and could cause very substantial gains to individuals by helping them manage their own thoughts and emotions to become happier. Given that mental illnesses have a sharply heighted suicide risk, Digital Campaigns, just like anti-poverty and anti-malaria programmes, are also in the business of trying to save lives.

However, it may be useful for me to defend my position even from theoretical objections. Voorhoeve has recently articulated a view called Aggregate Relevant Claims (ARC)[54] which tries to make this sort of objection precise, to which Halstead has recently responded, [55] and I should engage with these.

  • Reducing unhappiness has less value than increasing happiness

I also anticipate some people would think it’s much more important to reduce suffering rather than increase happiness, so Digital Campaigns to make happy people happier are unjustified. I think this is a similar mischaracterisation of my view: Digital Campaigns would take the negative mental states in mental illness and the negative mental states in ordinary human unhappiness, in addition to try to bring about more positive mental states.

Again, it may be useful to protect my view from the theoretical attack. I take it such an objection relies on a form of Negative Utilitarianism. The extreme version of this view is that only unhappiness matters, but all versions consider unhappiness as having greater disvalue than happiness has value. This view looks unpromising – as far as I can tell, it has no philosophical defenders[56] – because it implies the best thing to do would be to explode the world so that no one can suffer anymore.[57] It may be worth saying slightly more about this in the thesis.

  • Ignorance of the economic gains

This is the first of two practical objections. In this case, the worry is Digital Campaigns and Mental Health Charity might be more effective at promoting happiness over the short-term, but would not cause the same sort of long-run economic gains as anti-poverty and anti-malaria programmes. For example, suppose the billionaire ran a 6-month long poverty programme and a 6-month mental health programme, and only counted the effects, including on happiness, over those 6 months, the mental health campaigns might appear to cause comparatively more happiness than it does because the long-term economic gains are yet to materialise.

I think there is a practical and a methodological response to this. As it happens, it seems that mental health issues are a substantial economic drain. A first pass at the evidence shows they are associated with a loss of employment, absenteeism, poor performance within the workplace and premature retirement.[58] Mental illnesses also tend to have other co-morbidities which make them an economic cost to health systems.[59] The methodological response is that the billionaire should measure the impact of any programme over the long term, using happiness metrics as well conventional economics ones, such as unemployment, and be aware of the epistemic limits of his methods. I’d need to explore this area more and try to make sense of the follow on effects. I plan to take account of the recent accusations collated by Iason Gabriel that effective altruism suffers from methodological blindness.[60]

  • Making people happier could prolong misery

The second practical worry is teaching people to be happy may backfire. The scenario I am imaging is one where making someone happier causes people to stay in conditions they otherwise would not. [61] A serious criticism has already been made of effective altruism that the charity of donors from developed nations can sometimes have a soporific effect on the citizens of developing nations which stops them from taking action themselves.[62] Put most starkly, charity interventions keep corrupt elites in power by providing assistance that citizens would otherwise demand from those elites.[63]

I’d need to spend some time in my thesis exploring these criticisms. In particular, I’d want to understand if these objections are as pressing against EHIs as IHIs. Intuitively, I would think removing mental health conditions from people would enable them to pursue their own interests. This also raises the strange implication that it might be good to make people unhappy now to spur them into action later.

  • Other causes already more good

Not all utilitarians think they can do the most good by focusing on present humans. For those that focus on animal suffering, presumably they think the interventions we use for animals are X times more effective than our present EHIs for humans. Unless the new IHIs are X times more effective than the current EHIs, they are unlikely to change their position. I expect those concerned by the far, far future will still think making present human happier is effectively irrelevant.

  • Objections from those who object to utilitarianism’s approach to aggregation

I think it’s worth noting that, until this point, utilitarian, prioritarian and egalitarian effective altruists have been in almost total agreement that the right course of action was to fund anti-poverty and anti-malaria EHIs because those helped the worst off people in the world. [64] The utilitarian thought this was the best thing to do because the poorest were the most efficient at converting resources into happiness. If this conclusion is true for the utilitarian, it is more true, if anything, for the other two views. There are quite a lot of ways of spelling out prioritarianism and egalitarianism, but the general gist is that prioritarians believe welfare has diminishing value,[65] so helping the worse off is particularly valuable, and egalitarians think inequality has intrinsic disvalue,[66] so raising up the worse off is, again, particularly valuable. Hence effective altruists of all stripes largely agreed what to do.

On my account, we may find the effective altruists find themselves committed to different interventions on the basis of their differing accounts of the relationship between individual well-being levels and the overall goodness of states of affairs. At present, it’s not possible to say what non-utilitarians will want to do instead, for two reasons. First, all plausible moral views value the increase of happiness, so we need to know how cost-effective the various IHIs are. Second, we need to know how much happiness other views are prepared to trade off to achieve their preferred outcomes. For instance, if it turns out Digital Campaigns are 1000 times more cost effect at creating happiness than anti-malaria programmes, egalitarians and prioritarians are unlikely to think it would be worse to fund Digital Campaigns. I’m not going to discuss the relative merits of prioritarianism and egalitarianism to utilitarianism here although I will do so in the thesis.

(I also note that, if it turns out Digital Campaigns are more effective in developed countries, it may no longer be the case that MacAskill’s rule of thumb[67] – your donation does 100 time more good if spent in developing countries – no longer holds.)

  1. Conclusion

In this paper I’ve argued effective altruism, where it tried to increase the happiness of current humans, has so far focused exclusively on what I called ‘external happiness interventions’, which aim to increase happiness by changing people’s objective circumstances. I’ve said this overlooks internal and temporal happiness interventions – respectively, those which change how people think and spend their time. I then argued the empirical information on happiness meant internal happiness interventions looks very promising and outlined how they might be trialled. I concluded a classical utilitarian billionaire would do the most good with his money by trialling Digital Campaigns targeting mental health and ordinary human unhappiness around the world as well as funding innovation in Mental Health Charities in the developing world. Finally, I outlined some objections I did not have space to discuss.

 

 

 

 

 

 

Annex A[68]

Annex B

Below I generate some numbers compared the potential size of gains to happiness that could be made from removing malaria, poverty, mental illness and ordinary human unhappiness. On a hypothetical 10-point scale, I’ve estimated maximum and minimum amounts of happiness I believe someone could gain as a result of successful interventions, then multiplied by the number affected in any given year. I present the numbers, with the explanations for why I chose them given afterwards.[69] The 10 point scale is somewhat irrelevant: I could have just guessed at their comparative badness in a ratio scale with each other.

 

Number affected (millions) Potential happiness gain/person (10 point scale) Points of happiness gained (min) (millions) Points of happiness gained (max) (millions)
Ordinary human unhappiness 7125 0.5 – 2 3562.5 14250
Mental health 753.4 2 – 5 1506.8 3767
Malaria 146 0.5 – 1.5 73 219
Poverty 702 0.25 – 1.5 175 1050
Min gain from IHIs 5069.3 Min gain, M + P 248.5
Max gain from IHIs 18017 Max gain, M + P 1272
IHIs/M+P, Maximum difference 72.5
IHIs/M+P, Minimum difference 4.0

These numbers are highly speculative and provide an almost certainly wrong prediction to work from. As you can see, despite the wide range depending on assumptions about minimum and maximum values, direct attempts to improve human happiness is several times larger than poverty and malaria combined even by our most conservative estimates. In fact, even if you discount OHU altogether, the minimum estimated burden of mental health seems larger than the maximum burdens of malaria and poverty together.
To explain the numbers, as most people rate their subjective well-being at 7 on a 10-point scale,[70] my guess is we might be able to achieve a 2-point reduction in ordinary human unhappiness, with 9/10 being an effective limit (you couldn’t be maximally happy all the time). However, one might think the difference to be more modest (0.5/10), even in an ideal world. Poverty could have a very small impact on happiness (0.25/10). However, you might disagree with the earlier analysis and believe poverty could have nearly as big an impact on daily happiness as the average mental health disorder does (a maximum 1.5/10 for poverty and a minimum 2/10 for MH). This seems unlikely though: all unhappy people are unhappy, but not all poor people are unhappy. I believe mental health could be very bad (5/10). Regarding malaria, I guess that whilst it can be acute, it’s not chronic, like poverty, so is likely not worse than poverty.

I’ve not included deaths, because the analysis is too complicated and unlikely to add very much. For instance, malaria kills people, as does poverty, but it’s hard to work out what percentage of the deaths of the poor can be attributed to their poverty. Equally, mental health comes with a heightened suicide risk, and it seems being happier may help people live longer.

[1] Singer, P. (2015). The most good you can do: How effective altruism is changing ideas about living ethically

[2] MacAskill, W. (2015). Doing good better: Effective altruism and a radical new way to make a difference

[3] I suppose this could concern sentient life too, but in practice the focus seems to be on ensuring the survival of the human species.

[4] From Greaves, H. (2015) Antiprioritarianism Utilitas, 27(01), p.1

[5] I would have discussed the nature of happiness at length in an earlier thesis chapter

[6] Gabriel, I. (2016). Effective altruism and its critics. Journal of Applied Philosophy, p2

[7] MacAskill, W. (2015). Doing good better: Effective altruism and a radical new way to make a difference

[8] Givewell and Giving What We Can, effective altruist meta-charities (a charity that evaluations other charities) have their top recommendations available here http://www.givewell.org/charities/top-charities and https://www.givingwhatwecan.org/top-charities/. The two disagree slightly, but that is not important for this essay.

[9] On World Bank figures, press release available at http://www.worldbank.org/en/news/press-release/2015/10/04/world-bank-forecasts-global-poverty-to-fall-below-10-for-first-time-major-hurdles-remain-in-goal-to-end-poverty-by-2030

[10] Global Burden of Disease Study 2013 Collaborators. (2015). Global, regional, and national incidence, prevalence, and years lived with disability for 301 acute and chronic diseases and injuries in 188 countries, 1990-2013: A systematic analysis for the global burden of disease study 2013. Lancet (London, England), 386(9995), 743-800

[11] World Bank Press release, available at http://www.worldbank.org/en/news/press-release/2015/10/04/world-bank-forecasts-global-poverty-to-fall-below-10-for-first-time-major-hurdles-remain-in-goal-to-end-poverty-by-2030

[12] Global Burden of Disease Study 2013 Collaborators. (2015). Op. Cit.

[13] Analysis of AMF and Give Directly available at http://www.givewell.org/international/top-charities/

[14] They aren’t not mentioned in either Singer’s or MacAskill’s recent books, and meta-charity Giving What We Can’s post on the topic argues there are no cost-effective mental health interventions, available at https://www.givingwhatwecan.org/report/mental-health/

[15] A common alternative to ‘mental disorders’ is ‘MNS disorders’ (mental, neurological and substance abuse disorder). I don’t discuss neurological and substance abuse disorders as that requires a separate approach.

[16] Mental illness figures compiled from the Global Burden of Disease Study 2013 Collaborators. (2015). Op. Cit.

[17] Organisation for Economic Co-operation and Development. (2013). OECD guidelines on measuring subjective well-being

[18] There’s sometimes also a ‘eudaimonic’ component, which measure how meaningful the person thinks their life to be, which isn’t relevant to discuss here

[19] Kahneman, D., & Deaton, A. (2010). High income improves evaluation of life but not emotional well-being. Proceedings of the National Academy of Sciences of the United States of America, 107(38), 16489-16493. doi:10.1073/pnas.1011492107 [doi]

[20] Diener, E., Lucas, R. E., & Scollon, C. N. (2009). Beyond the hedonic treadmill: Revising the adaptation theory of well-being. The science of well-being (pp. 103-118) Springer.

[21] Wilson, T.D. and Gilbert, D.T., 2005. Affective forecasting knowing what to want. Current Directions in Psychological Science, 14(3), pp.131-134.

[22] Haushofer, J. et al. (2015) Your Gain Is My Pain: Negative Psychological Externalities of Cash Transfers, Working Paper.

[23] Suggested by Hilary Greaves in private conversation

[24] Easterlin, R.A. and Angelescu, L., 2009. Happiness and growth the world over: Time series evidence on the happiness-income paradox.

[25] Kahneman, D. and Deaton, A., 2010. High income improves evaluation of life but not emotional well-being.

[26] Oswald, A.J. and Powdthavee, N., 2008. Does happiness adapt? A longitudinal study of disability with implications for economists and judges. Journal of public economics, 92(5), pp.1061-1077.

[27] Schulz, R. and Decker, S., 1985. Long-term adjustment to physical disability: the role of social support, perceived control, and self-blame. Journal of personality and social psychology, 48(5), p.1162

[28] Wilson, T.D. and Gilbert, D.T., 2003. Affective forecasting. Advances in experimental social psychology, 35, pp.345-411.

[29] Frey, B.S. and Stutzer, A., 2014. Economic consequences of mispredicting utility. Journal of Happiness Studies, 15(4), pp.937-956

[30] National Institute for Clinical and Health Excellence. Guide to the Methods of Technology Appraisal. London: National Institute for Clinical and Health Excellence; 2008.

[31] Dolan P. Modelling valuations for EuroQol health states. Med Care. 1997;35:1095–108.

[32] Dolan, P. and Metlcalfe, R., (2012). Valuing Health A Brief Report on Subjective Well-Being versus Preferences. Medical decision making, 32(4), pp. 579

[33] Ibid. pp579

[34] Ibid. p580

[35] Ibid. p580

[36] Information from the Centre for Disease Control, available at: http://www.cdc.gov/malaria/about/disease.html

[37] Hofmann, S.G et al (2012) The efficacy of cognitive behavioural therapy: a review of meta-analyses. Cognitive therapy and research, 36(5), pp.427-440.

[38] Khoury, B., et al (2013). Mindfulness-based therapy: A comprehensive meta-analysis. Clinical Psychology Review, 33(6), pp.763-771.

[39] Seligman, M.E., Rashid, T. and Parks, A.C., 2006. Positive psychotherapy. American psychologist,61(8), p.774.

[40] Cukrowicz, K.C. and Joiner Jr, T.E., 2007. Computer-based intervention for anxious and depressive symptoms in a non-clinical population. Cognitive Therapy and Research, 31(5), pp.677-693

[41] Kaviani, H., Javaheri, F. and Hatami, N., 2011. Mindfulness-based cognitive therapy (MBCT) reduces depression and anxiety induced by real stressful setting in non-clinical population. International Journal of Psychology and Psychological Therapy, 11(2), pp.285-296

[42] Sin, N.L. and Lyubomirsky, S., (2009). Enhancing well‐being and alleviating depressive symptoms with positive psychology interventions: A practice‐friendly meta‐analysis

[43] Kaltenthaler, E., Parry, G., Beverley, C. and Ferriter, M., 2008. Computerised cognitive–behavioural therapy for depression: systematic review. The British Journal of Psychiatry, 193(3), pp.181-184

[44] http://www.givewell.org/international/top-charities/amf#footnoteref119_eec94dw

[45]Patel, V et al. (2015). Addressing the burden of mental, neurological, and substance use disorders: key messages from Disease Control Priorities. The Lancet. These figures should be treated with caution. As the authors note, p1681 “hardly any published evidence exists on the cost-effectiveness of population-based or community-level strategies in or for low-income and middle-income settings”

[46] Lower Middle Income Countries can be understood as ‘developing countries’ and refers to those on the following list: http://data.worldbank.org/income-level/LMC

[47] Saxena, S., A. Lora, et al. (2007). “WHO’s Assessment Instrument for Mental Health Systems: collecting essential information for policy and service delivery.”

[48] World Health Organization (2011). Mental Health Atlas 2011. Geneva

[49] de Menil, V., 2015. Missed Opportunities in Global Health: Identifying New Strategies to Improve Mental Health in LMICs.), p8

[50] Ibid, p1

[51] From Gates’ Foundations website, available at: http://www.gatesfoundation.org/What-We-Do

[52] Gabriel, I. 2016. Effective altruism and its critics. Pp11

[53]Scanlon, T., 1998. What we owe to each other. p236

[54] Voorhoeve, A., 2014. How Should We Aggregate Competing Claims? Ethics, 125(1), pp.64-87.

[55] Halstead, J., 2016. The Numbers Always Count*. Ethics, 126(3), pp.789-802.

[56] As pointed out by Toby Ord, 2013 “Why I am not a negative utilitarian” available from http://www.amirrorclear.net/academic/ideas/negative-utilitarianism/

[57] First pointed on in R. N. Smart. ‘Negative Utilitarianism’, Mind 67:542–3. 1958

[58] McDaid et al. 2005. Policy Brief: Mental Health III: Funding mental health in Europe. Copenhagen: World Health Organization and European Observatory on Health Systems and Policies

[59] Miller, B. J. and Paschall, C. B., 3RD & SVENDSEN, D. P. 2006. Mortality and medical comorbidity among patients with serious mental illness. Psychiatr Serv, 57, 1482-7.

[60] Highlighted in Gabriel, I. 2016. Effective altruism and its critics. pp6

[61] In conversation Julian Savalescu suggested teaching an abused partner to be happier may have this effect.

[62] See Gabriel, I., 2016. Effective Altruism and its Critics.pp12

[63] For example, see Emily Clough’s argument in the Boston Review, available https://bostonreview.net/world/emily-clough-effective-altruism-ngos

[64] Ibid, 2-6 Gabriel argues ‘thick’ effective altruists do sometimes reach slightly different conclusions from prioritarians, egalitarians and advocates of rights.

[65] Derek Parfit, ‘Equality and priority’, Ratio 10,3 (1997): 202 – 221, at p. 213.

[66] Larry Temkin, Inequality (Oxford: Oxford University Press, 1993), p. 13

[67] MacAskill 2015 op. cit., p. 28.

[68] From Dolan, P. and Metlcalfe, R., (2012). Valuing Health A Brief Report on Subjective Well-Being versus Preferences. Medical decision making, 32(4), pp. 578-582

 

[69] These numbers were originally guessed at in an unpublished draft paper I wrote with Konstantin Sietzy about effective altruism and mental health.

[70] For instance, see of the Office of National Statistics measures of subjective well-being from the UK, available at http://www.ons.gov.uk/peoplepopulationandcommunity/wellbeing/bulletins/measuringnationalwellbeing/2015-09-23#summary for the sake of this analysis I don’t think it matters all the numbers are evaluations, rather than experiences, of happiness.