Which cause is most effective?

Introduction

In previous posts, we explained what causes are and presented a method for assessing them in terms of expected effectiveness.

In this post, we apply this method to identify a list of causes that we think represent some particularly promising opportunities for having a social impact in your career (though there are many others we don’t cover!).

We’d like to emphasise that these are just informed guesses, over which there’s disagreement. We don’t expect the results to be highly robust. However, you have to choose something to work on, so we think it’ll be useful to share our guesses to give you ideas and so we can get feedback on our reasoning – we’ve certainly had lots of requests to do so. In the future, we’d like more people to independently apply the methodology to a wider range of causes and do more research into the biggest uncertainties.

The following is intended to be a list of some of the most effective causes in general to work on, based on broad human values. Which cause is most effective for an individual to work on also depends on what resources they have (money, skills, experience), their comparative advantages and how motivated they are. This list is just intended as a starting point, which needs to be combined with individual considerations. An individual’s list may differ due also to differences in values. After we present the list, we go over some of the key assumptions we made and how these assumptions affect the rankings.

We intend to update the list significantly over time as more research is done into these issues. Fortunately, more and more cause prioritisation research is being done, so we’re optimistic our answers will become more solid over the next couple of years. This also means we think it’s highly important to stay flexible, build career capital, and keep your options open.

Some further qualifications:

  • We’re not presenting our full reasoning in this post. That would take up too much space. Rather, we intend to write more about each individual cause as they arise in case studies.
  • The list is just some promising opportunities, and is not comprehensive. There are many causes we haven’t even been able to consider.
  • There is much variation within causes: an unpromising cause can contain a highly promising intervention and promising causes can contain useless interventions. We still think it’s useful to organise your career around a cause, but it’s important to remember that if an organisation supports a promising cause, it doesn’t guarantee the organisation is effective, and a cause being of low priority doesn’t rule out the organisation from being highly effective.
  • When we name organisations within each cause, do not take this as an endorsement of this organisation. If an organisation’s working on a high priority cause, we think that’s a point in the organisation’s favour, but the organisation can easily fail to be effective.
  • These causes are assessed from a global perspective. We don’t investigate which causes are most effective for helping your local community.
  • The ratings ‘2’, ‘3’, ‘4’, etc. are just meant as a relative assessment of the factor for this cause compared to the others. The numbers do not correspond to any scale.

In the rest of this post we:

  1. Provide a summary list of high-priority causes
  2. Explain what each cause is and overview our reasons for including it
  3. Explain how key judgement calls alter the ranking
  4. Overview how we came up with the list and how we’ll take it forward
  5. Answer other common questions

The List

This is the list produced by applying our cause framework to the most promising causes we currently know. Do not read too much into the order of the list – it’s highly dependent on assumptions, which we’ll overview later.

‘5’ represents for ‘very high, relative to the others’, ‘3’: ‘average, relative to the others’, and ‘1’: ‘low, relative to the others’.

CauseImportanceTractabilityUncrowdedness
Prioritisation research 545
Promoting effective altruism 545
Global catastrophic risks 4 25
Research policy and infrastructure 33.53
Ending factory farming 244
Global health 253
Improving decision making 333
Immigration reform 324
Geoengineering research 234
Biomedical research 242
Developing world economic empowerment 322

Would you like to comment on these scores? Go here.

Skip ahead to read more about each cause:

  1. Prioritisation research
  2. Promoting effective altruism
  3. Global catastrophic risks
  4. Research policy and infrastructure
  5. Ending factory farming
  6. Global health
  7. Improving decision making
  8. Immigration reform
  9. Geoengineering research
  10. Biomedical research
  11. Developing world economic empowerment

Why these causes?

Prioritisation Research

What is this cause?
Prioritisation research is activity aimed at working out which causes, interventions, organisations, policies, etc. do the most to make the world a better place. Organisations and projects within this cause include some policy think-tanks and some parts of economics. Within prioritisation research, we think the most high-priority area is long-run-focused cause-prioritisation. That is, research aimed at working out which causes do the most to make the world a better place in the long-run if we add more resources to them. Note that this research need not consist of detailed economic modelling. Cause-prioritisation can also involve down-to-earth projects like investigating room for more funding or aggregating expert opinion. Organisations within this sub-cause include the Copenhagen Consensus, GiveWell, the Future of Humanity Institute and the Centre for Effective Altruism (our parent charity).

Why do we think it’s high-priority?

(See our cause framework for more explanation of each factor)

Important: 5
Tractable: 4
Uncrowded: 5

We think cause-prioritisation is a highly effective cause, because: (i) we think there are likely to be large differences in the effectiveness of different causes, (ii) many don’t have a good understanding of these differences, and (iii) without a better understanding, we are unlikely to take the best opportunities to do good. We also think working on this cause offers high value of information. Since there hasn’t been a large systematic attempt to evaluate causes before, even if the project turns out not to produce useful answers, it’ll still be highly useful to rule it out as a promising project.

At the same time, we think cause-prioritisation is tractable and uncrowded. Little has been directly spent on this kind of project so far – there are only three major organisations working on the cause and their annual budgets are under US$2m – but there are hundreds of billions of dollars at stake in philanthropy and government aid spending. What has been spent so far, however, has led to significant progress, for instance GiveWell identifying global health as a promising area to look for effective donation opportunities, the Copenhagen Consensus’s promotion of micronutrient supplements, and the development of better methodologies for prioritisation (e.g. how to make use of cost-effectiveness estimates). Moreover, there are promising lines of future research, and organisations within the cause that are short of funding and human capital (see here, here, and here).

One important weakness of this cause is that, as with many research programs, it can be difficult to tell when you’re making progress, which lowers tractability.

We’d like to flag that there are reasons we may be biased. Our parent charity, CEA, supports cause-prioritisation. Moreover, the creation of this list is itself an exercise in cause-prioritisation, so you might expect us to rate it highly. On the other hand, our high rating of cause-prioritisation is not an accident. CEA and 80,000 Hours aim to work on the most high-potential causes in the world. It’s because we think cause-prioritisation is a high-priority cause that we’re working on it (our money is where our mouth is). So, although it’s true there’s potential for a conflict of interest, there’s a good reason we’re in this situation. The greater risk is in the future: we’re likely to be biased towards continuing to believe prioritisation research is high-priority because we’re already working on it (an instance of the sunk cost bias) despite new evidence potentially suggesting otherwise. We’ll attempt to guard against this risk.

You can see much more detail on the case for cause prioritisation in this draft report. We’re planning a more thorough overview of what opportunities are available within the cause.

We also think that other types of prioritisation research are high-priority (e.g. charity evaluation), but they seem less important and less uncrowded compared to cause-prioritisation research.

See all of our resources on prioritisation research

Promoting effective altruism

What is this cause?
Promoting effective altruism means activities which expand the capabilities of those trying to do good in a cause-neutral, evidence-based and outcome-orientated way. Interventions within this area include advocacy of key ideas in effective altruism and network-building. Some organisations in this cause include GiveWell, the Centre for Effective Altruism (our parent charity), the Copenhagen Consensus, Leverage Research, Charity Science, the donation pledge organisations (Giving What We Can, The Life You Can Save, the Giving Pledge) and ourselves. More broadly, you could also include organisations with an effectiveness-minded approach, like the Gates Foundation and Evidence Action.

How is it different from prioritisation research?
Prioritisation research is working out which opportunities have the most impact, while promoting effective altruism is building capacity to act on this research. In practice, both need to be carried out at the same time, and many organisations engage in a mixture of both.

Why do we think it’s high-priority?

Important: 5
Tractable: 4
Uncrowded: 5

Promoting effective altruism is effective because it’s a flexible multiplier on the next most high-priority cause. It’s important because we expect the most high-priority areas to change a great deal, so it’s good to build up general capabilities to take the best opportunities as they are discovered. Moreover, in the recent past, investing in promoting effective altruism has resulted in significantly more resources being invested in the most high-priority areas, than investing in them directly. For instance, for every US$1 invested in GiveWell and Giving What We Can, more than $7 have been moved to high-priority interventions. We think it’s highly important also because it’s a brand new area with high potential, so we expect further work to have high value of information.

Promoting effective altruism seems uncrowded, because it’s a new cause so there appear to be lots of good opportunities within it which haven’t been taken yet. It seems tractable because there are definite advocacy opportunities, which have worked in the past and whose success can be measured, e.g. encouraging people to take the GWWC pledge. More direct evidence for effectiveness comes from the strong success to date of many of the projects in the area, like GiveWell.

One important weakness of this cause is that, as with most advocacy projects, it’s difficult to be confident that the interventions which have worked in the past will continue working into the future. This lowers tractability. There are also reasons effective altruism might fail to be a good project – see here for some ideas.

Again, there’s reason for us to be biased. 80,000 Hours is involved in promoting effective altruism, so it’s in our interests to say this is a high-priority cause. However, our prioritisation of promoting effective altruism is no accident. Our aim is to work on the most high-priority causes in the world. We set up 80,000 Hours precisely because we think promoting effective altruism is a high-priority cause, so it’s no surprise we rank it highly. The greater risk is in the future: we’re likely to be biased towards continuing to believe prioritisation research is high-priority because we’re already working on it (an instance of the sunk cost bias) despite new evidence potentially suggesting otherwise. We’ll attempt to guard against this risk.

See all our resources on promoting effective altruism.

Global catastrophic risks

What is this cause?
Working on global catastrophic risks means identifying and mitigating low probability but costly risks to society’s future. Examples of interventions in this category include setting up early warning systems for natural disasters, surveillance of pandemics, tracking asteroids (which has largely been completed for asteroids that threaten civilization, although not for comets), advocacy of non-proliferation of nuclear weapons, and research on other possible risks and methods for mitigating them. An unusual view we take seriously is that some of the most significant risks in this area will come from new technologies that may emerge this century, such as synthetic biology, distributed manufacturing, or artificial general intelligence (which we often call ‘unprecedented risks’). Organisations within this cause include various parts of government (e.g. DARPA, NASA), various think-tanks (e.g. those working on nuclear weapon risk), small parts of the insurance industry, the Global Catastrophic Risks Institute, the Nuclear Threat Initiative, the Skoll Foundation, the Sloan Foundation, the Machine Intelligence Research Institute, the Centre for the Study of Existential Risks at Cambridge and the Future of Humanity Institute at Oxford.

Why do we think it’s high-priority?

Important: 4
Tractable: 2
Uncrowded: 5

Progress in this area has a clear link with better long-run outcomes for society – catastrophic risks cause huge damage and put society’s long-term future in danger. So we think this cause is highly important. Past lack of effort invested in many of these risks, and high uncertainty about how to mitigate them, also means we can expect high value of information from working on this cause. We don’t rate it as ‘5’, however, because we’re unsure of the importance of working on mitigating these risks directly, compared to working on other causes which also have the potential to benefit the long-run future, either through reducing catastrophic risks or through other means.

This cause seems uncrowded because the risks considered relate to potential harm in the future, and there’s good reason to expect present society to undervalue the interests of future generations. It also seems that irrational biases discourage people from working in this area (e.g. undervaluing of small risks that haven’t occurred before).1 There’s more direct evidence for uncrowdedness in the fact that top foundations only spend 0.1% of their resources on these risks, which seems small relative to what’s at stake.

The main weakness of this cause is tractability – there’s a huge amount of uncertainty about which interventions will effectively reduce these risks in the future. For instance, doing more research into the risks posed by synthetic biology could accidentally further the discovery of a dangerous application of synthetic biology. On the other hand, there are some interventions within this cause which seem relatively straightforward, like preparing better systems for managing disasters and performing better tracking of the people developing potentially dangerous technology. There have also been some good interventions within this cause in the past, such as asteroid tracking. See this rough analysis of the cost-effectiveness of asteroid tracking efforts to date, and GiveWell’s overview of asteroid tracking, which found it was promising but doesn’t currently have much room for more funding. Overall, however, due to lower tractability and probably lower importance, we currently prefer further prioritisation research and general capacity-building through promoting effective altruism.

For more, see Global Catastrophic Risks by Nick Bostrom, Our Final Century by Lord Martin Rees, GiveWell’s shallow overviews of several sub-causes in this area, and conversation notes, and see all of our resources on this cause.

Research policy and infrastructure

What is this cause?
Research policy and infrastructure is activity aimed at increasing the extent to which scientific research benefits society. Interventions within this cause include: promoting systematic reviews (e.g. Cochrane Collaboration), campaigns to enforce pre-registration of trials, replication projects, improving methods within specific fields (e.g. promoting the use of randomised controlled trials in development studies, as pursued by JPAL), and developing platforms to promote open science (promoting new ways to produce, share and evaluate scientific research, advanced by organisations like Academia.edu, Mandeley, Digital Science, and the Open Science Foundation). The Arnold Foundation is a large supporter of this cause.

Why do we think it’s high-priority?

Important: 3
Tractable: 3.5
Uncrowded: 3

Scientific research is hugely important. It has driven much of our improvement in living standards in recent history, and is a major driver of long-term productivity (a couple of examples among many: smartphones, advances in HIV treatment, improvements in crop yields). However, it seems there could be considerable room to make some areas of science more efficient (e.g. see Bad Pharma by Ben Goldacre for problems with medical research, and see GiveWell’s overview of some of the problems in this field). The cause seems tractable because a variety of concrete proposals for improving the effectiveness of science are on the table. It also seems moderately uncrowded. We can expect basic scientific research in general not to receive enough investment because it’s difficult to capture the benefits for oneself, which will mean it’s undersupplied by the market. In addition, the benefits mostly accrue in the long-term future, so present people are under-incentivised to invest in it. Within this, research policy and infrastructure seem particularly neglected because most of the key players in scientific research do not seem incentivised or well-placed to promote it. It’s not part of traditional academic research, but it requires more scientific expertise than policy makers or businesses can easily provide.

On the other hand, this cause seems more crowded than those above. GiveWell’s major investigation into open science (which they initially thought was the most promising sub-cause for donors) showed that there were major attempts by for-profit businesses to solve the problems. There has also been more progress to date, which makes interventions at the margin seem less tractable than the other causes.

See all our resources on research policy and infrastructure.

Ending factory farming

What is this cause?
Ending factory farming is activity aimed at stopping the practice of animals being raised in suffering for food. Broadly, the interventions in this area are advocacy aimed at reducing how much factory farming takes place, or research aimed at determining the most effective methods of advocacy, developing meat substitutes or higher-welfare farming methods. Some organisations in this cause include Animal Charity Evaluators, the Humane League, Vegan Outreach, Beyond Meat and New Harvest.

Why do we think it’s high-priority?

Important: 2
Tractable: 4
Uncrowded: 4

Ending factory farming is important because animals suffer in huge numbers in factory farms. Around 60 billion animals are raised for food each year, the majority of which are in factory farms. Moreover, the cause seems tractable and uncrowded because there is some evidence that advocacy campaigns aimed at encouraging vegetarianism reduce the number of factory farmed animals for a very low cost. Developing meat substitutes could also provide a high-leverage way to reduce consumption of factory-farmed meat. Further, we can expect the cause to be undervalued because the interests of animals are not well-represented by our economic or political system. Even among charity, less than three percent of US donations in 2011 went to the “environment/animals” sector, which includes zoos, aquariums, and programs for “outdoor survival and beautification of open spaces”. Only a tiny portion of that went to animal charities, and within animal charities, the majority of attention is given to animal shelters rather than factory farming. We can also expect further work to have high value of information, because it seems relatively little is known about the most promising efforts within this cause. Finally, we take it seriously because a significant number of effective altruists who have thought about how to do the most good rate it as the most high-priority cause.

If you attach importance to the long-run future, then the main disadvantage of acting against factory farming is that there doesn’t seem much reason to expect that reducing factory farming is among the best ways to contribute to a generally flourishing future in the long-term. Reducing factory farming can produce some long-term flow-through effects through reducing crop prices, carbon emissions and promoting anti-speciesism. Reducing crop prices and carbon emissions are reasons the Gates Foundation supports research into in vitro meat as a way to further global development. Nevertheless, if you want to promote these outcomes, it seems unlikely to us that working to reduce factory farming is the most effective way to go about it. We also don’t see strong reason to think that reducing crop prices and carbon emissions are among the best ways to promote general global development. For instance, Giving What We Can recently concluded that it’s currently likely to be significantly more effective to work on global health, than reducing carbon emissions. Nevertheless, there hasn’t been significant research into these issues, so we could easily change our mind.

For more, see GiveWell’s overview of the cause and all our other resources on this cause.

Global health

What is this cause?
Global health is activity to reduce the incidence of illness globally, and particularly in the developing world. We can improve global health through several broad avenues, including biomedical research (which is also treated separately below), improving public health, and promoting international aid. Some of the more promising projects within global health include expanding the availability of insecticide-treated bed-nets, deworming, developing vaccines for HIV and neglected tropical diseases, further cost-effectiveness research, and increasing the cost-effectiveness of existing aid and philanthropy. There are many organisations within this cause, including the World Bank, World Health Organisation, the Gates Foundation, a large number of medical research bodies, and all governments.

Why do we think it’s high-priority?

Important: 2
Tractable: 5
Uncrowded: 3

Global health is important because health really matters to our wellbeing and productivity, yet millions of people suffer from ill health. The best thing about global health as a cause is that plenty of highly tractable interventions exist that could easily be expanded if more resources were added to the cause. For instance, interventions like insecticide-treated bed-nets have been shown by multiple randomised controlled trials to significantly reduce the burden of malaria, for very low costs (several thousand dollars per life saved). Most of the Copenhagen Consensus 2012’s top-ranked interventions were within global health (e.g. the top three interventions were micronutrients to school children, subsidy for malaria combination treatment, and expanded childhood immunisation coverage). An additional benefit is that the impact of health interventions is relatively easy to quantify, which makes it easier to select the best programs and learn from failure.

If you attach importance to the long-run future, then the weakness of global health is that although we’re highly confident that the short-term impact is highly positive, we know very little about whether improving global health is a particularly promising way to develop a flourishing society in the long-run. Moreover, because it’s relatively well-explored, we don’t expect additional work to have particularly high value of information. Both of these problems reduce the importance score. The cause also receives significant attention from major strategic actors (like the Gates Foundation), and although we’re confident there are good opportunities for donors in the cause, we’re less sure how talent-constrained it is (particularly because it’s a fairly popular cause), so we think it does less well on crowdedness.

For more, see GiveWell’s arguments in favour of global health as the top cause for donors. Also see GiveWell’s conversations within this cause, and their overviews of some relevant interventions. See all our resources on global health.

Improving decision-making

What is this cause?
Improving decision-making means improving our ability to form accurate beliefs about the world and act on this information to achieve our goals. This is a broad cause, including a growing research program aimed at improving forecasting, for instance Philip Tetlock’s Good Judgement Project, studies of expert judgement in psychology and behavioural economics (see Thinking Fast and Slow by Kahneman for an overview), prediction markets (e.g. as promoted by Robin Hanson), and efforts to develop rationality training, as advanced by the Center for Applied Rationality.

Why do we think it’s high-priority?

Important: 3
Tractable: 3
Uncrowded: 3

Improving decision-making is rated highly as a cause by many in the effective altruism community, including those at the Center for Applied Rationality. It’s an important cause because if we improve our general abilities to achieve our goals, then we can expect the world to be made a better place without knowing the details of what’s going to happen in the future. In particular, improved decision-making could increase society’s ability to deal with a variety of important global challenges (including global catastrophic risks). We don’t rate it as ‘4’ however, because we haven’t seen strong evidence to show that it’s more important than other types of general empowerment or more direct approaches like working directly on understanding catastrophic risks.

The cause seems uncrowded, at least in some parts. For instance, the Centre for Applied Rationality is the only organisation we know to be working on rationality training. On the other hand, there are major research programs in psychology and economics working on some issues within this cause, so we don’t think it’s highly uncrowded. We also haven’t been presented with evidence that there’s a particularly pressing need for more resources within this cause.

We rate tractability ‘3’, because although some approaches to improving decision-making have been identified, we haven’t seen much evidence to suggest they’ll be effective to implement in practice, or that they would have a large impact on our lives.

See all our resources on improving decision making.

Immigration reform

What is this cause?
Immigration reform is advocacy of loosening immigration restrictions in rich countries with stronger political institutions, especially for people who are migrating from poor countries with weaker political institutions. It also includes research aimed at analysing how to effectively implement immigration reform, which seems particularly high-priority given the uncertainties and fears around the potential harmful side effects of increased immigration. Some organisations in this cause include the Center for Global Development, Fwd.us and the Krieble Foundation.

Why do we think it’s high-priority?

Important: 3
Tractable: 2
Uncrowded: 4

This cause is important because individual workers in poor countries could produce things of much greater economic value and better realise their potential in other ways if they lived in rich countries, meaning that much of the world’s human capital is being severely underutilised. This claim is unusually well-supported, by basic economic theory and the views of a large majority of economists. Immigration reform has the potential to yield a massive reduction in global poverty. For instance, remittances from migrants to their home countries are already twice as large as international aid, and this could be increased several fold. The cause seems uncrowded. Only four of the top 100 US foundations focus on it. It’s avoided by politicians because their constituents will not be the beneficiaries of the cause’s reforms – they’ll instead be members of the global poor. We don’t rate this cause more highly because many concerns have been raised around the political feasibility and social consequences of migration, which means that although increased migration is likely very beneficial in principle, we’re not sure which real interventions would have large positive effects.

For more, see GiveWell’s shallow overviews within the cause: here and here, and see all our resources on this cause.

Geoengineering research

What is this cause?
Geoengineering research is activity aimed at working out whether there are safe, effective ways to artificially alter the climate in order to prevent dangerous climate change. The main funding comes from governments and the Gates Foundation. People carry out the research within academic climate science.

Why do we think it’s high-priority?

Important: 2
Tractable: 3
Uncrowded: 4

Geoengineering research could be important because geoengineering may be a very cheap way to prevent dangerous climate change, but we’re highly uncertain, so more research would have high value of information. At the same time, this cause seems highly neglected. Geoengineering overall only receives about 0.1% of total spending on climate research.2 Solar geoengineering only receives around US$10m in funding per year – small relative to its potential importance – so it seems reasonable to expect further research to reduce our uncertainty about the benefits of geoengineering. The Copenhagen Consensus has made a rough benefit-cost estimate, showing high cost-effectiveness, and it has been a high-priority area for GiveWell Labs to investigate.

Note that whether geoengineering is a good idea is highly controversial among climate scientists. What we recommend in this cause is only further research into the benefits and costs of geoengineering. Some experts, however, even caution against further geoengineering research, since it may increase the chances that geoengineering is used inappropriately, so we’ve reduced the importance score. In addition, this research may prove to be highly intractable, due to the difficulties of modelling the climate.

See all our resources on geoengineering.

Biomedical research

What is this cause?
Biomedical research is research aimed at developing ways to improve health through current scientific means. The main types of interventions in this area include doing research, supporting research, and advocating better or more effective government spending. The main organisations in this cause are governments, universities, foundations like the Wellcome Trust, and the pharmaceutical industry.

Why do we think it’s high-priority?

Important: 2.5
Tractable: 4
Uncrowded: 2

Biomedical research is important because health is a highly important part of wellbeing, and there’s a great deal we could do to improve our health. Biomedical research can have both considerable short run effects by combating disease and positive long-run effects by helping to build the store of scientific knowledge. In addition, biomedical research includes work on some potentially transformative developments, including synthetic biology (making it possible to make designer viruses), ending ageing and embryo selection. However, we don’t rate importance more highly, because society has already invested a great deal in biomedical research and the space for improvement seems smaller with other causes. We also rate it lower for value of information for the same reason.

It’s tractable because there’s a large number of existing promising research programs that people can contribute to, which we expect to lead to more progress. It has also had a strong track record of success, yielding some of the most important advances in living standards over the last one hundred years. GiveWell performed a literature survey on the economic returns to biomedical research, finding some evidence for very high returns.

Biomedical research receives a lot of attention and considerable support from government, industry and philanthropy, so it’s not uncrowded. However, we might still expect that it doesn’t receive enough investment relative to its importance. Why not? First, as with any type of innovation, it’s difficult to capture the benefits of research, which is reason to expect it to be undersupplied by the market, especially for more fundamental and less applied research. Second, the payoffs from biomedical research arrive decades in the future, or even later, which means that present society is likely to be under-incentivised to invest in it. Third, many of the benefits (at least for some diseases) will primarily be aimed at the global poor, so the financial incentives are lower than they should be. Within biomedical research, there are likely to be some neglected opportunities. For instance, we’ve across some intuitively plausible arguments that anti-ageing research has particularly high potential, though we haven’t vetted these claims.

As further evidence, we’ve seen benefit-cost analysis produced by the Copenhagen Consensus to suggest that various types of biomedical research are highly cost-effective in expectation (e.g. HIV vaccine research), and it has been a significant research priority for GiveWell Labs.

For more, see all our resources on biomedical research.

Economic empowerment of the developing world

What is this cause?
Developing world economic empowerment is activity aimed at increasing the economic power and wealth of the global poor. It includes a wide range of activities, including efforts to increase crop yields, providing financial services to the global poor, cash transfers, providing training, increasing the ease of doing business, and making investments aimed at increasing economic output. There’s a huge number of organisations working for this cause, including the Gates Foundation, World Bank, Give Directly, and many major charities (e.g. Oxfam).

Why do we think it’s high-priority?

Important: 3
Tractable: 2
Uncrowded: 2

This cause is important because 2.5 billion people live on less than US$2 a day. These people lack many of the basic necessities of life, including food, water, shelter and sanitation.

The problem with this cause is that many of the interventions within it don’t have a track record of being highly effective. For instance, despite billions of dollars of investment, there isn’t much convincing evidence that microfinance (see GiveWell’s overview, or our own) has had an outsized economic impact. This form of aid has received some of the most criticism from economists, including William Easterly and Dambisa Moyo, who have argued that some economic aid has caused significantly more harm than good. It also strikes us as crowded, because it receives a large degree of attention from existing aid programs and NGOs. Overall, we think global health may be a more promising area if your aim is empowerment of the global poor, because there’s a wide variety of health interventions with good evidence for both significant health benefits and economic benefits.

On the other hand, there are some promising interventions within this cause. Give Directly is highly rated by GiveWell – since we are so much richer than the global poor, even simple cash transfers can yield significant benefits, with good evidence. The Copenhagen Consensus rates research into increasing crop yields highly, and the Gates Foundation supports a variety of interventions within this cause.

For more, see all our resources on developing world economic empowerment.

What are the most important judgement calls we made in constructing this list?

There are many difficult judgement calls behind our application of this methodology, which many people will disagree over. It’s tempting to think that the existence of these individual differences makes the entire project to generally prioritise causes a waste of time. We don’t think this is true. In practice, we think there is enough overlap in values that there’s a lot we can say about causes in general, especially for people aiming broadly to have a positive impact with a global perspective.

This list has been constructed with the assumption that what’s ultimately valuable is something like human welfare, today and in the long-run, and that all people are equally valuable. For more, see our cause framework. If you primarily only value your local community, friends and family, then this list isn’t going to be of much use. However, if you care about the global perspective to some degree, then we suspect it will be useful.

On the other hand, within this broad perspective there are still many difficult judgement calls that members of 80,000 Hours disagree about, which are important in determining the rankings. Our approach with these is to clearly flag them so that you can make up your own mind. We may also expand on certain specific topics if there’s enough demand. In the following section, we explore some of these judgement calls and provide alternative lists based on these different assumptions.

Moral judgements

The question of which causes are most important is in part a moral question. For instance, you’ll only think that the existence of factory farming is a problem if you think animal suffering is morally relevant (unless you’re pursuing it to reduce carbon emissions and crop prices). Other moral issues can also arise. For instance, you might think a cause could help more people have flourishing lives, but that we don’t have an immediate moral reason to do anything about it (e.g. actions aimed at helping people who don’t exist yet).

In this list, one key judgement call is the relative importance of helping future people. We think future people deserve moral consideration, which means that the impact of our actions on the future is highly important. If someone thinks future generations deserve less consideration, they may favour global health, immigration reform and ending factory farming. The list might start with: promoting effective altruism, ending factory farming, global health, prioritisation research and immigration reform.

There are many other moral judgements that may alter our list. For instance, if someone places significant moral weight on justice as an end in itself, then they may want to focus on improving global governance. Unfortunately, these kinds of trade-offs are highly unexplored.

Judgement calls about how the world is

The relative importance of the global catastrophic risks cause depends on how likely you think these events are, and how much we can do about them today or in the usefully near future. For each risk, it seems that experts often disagree about their likelihood and how best to mitigate them.

Another major judgement call over which some people disagree is whether we can expect the future to be ‘good’ or ‘bad’ overall. If someone were to think of this as an uncertainty, or that it’s likely to be bad, then they’ll be less concerned about some catastrophic risks, and will instead likely want to focus on ways to make the future better. This person’s list might start with: prioritisation research, promoting effective altruism, ending factory farming, improving decision-making, and research policy and infrastructure.

Relatedly, if someone thinks it may not be good to speed up technological or economic progress, then they may place less importance on global health, immigration reform, biomedical research and some research policy and infrastructure. Their list might instead start with: prioritisation research, promoting effective altruism, global catastrophic risks, ending factory farming, improving decision-making, geoengineering research and research policy and infrastructure.

Each individual cause involves many assumptions about how the world is. For instance, behind prioritisation research lies the assumption that causes vary significantly in effectiveness, and that it’s possible to make progress in working out which are best. Behind promoting effective altruism lies the assumption that effective advocacy methods will continue to exist, and that a more effectiveness-minded approach is significantly better than what already exists.

Judgement calls about what we can know, based on the evidence

Some people differ over how generally sceptical to be about finding particularly good interventions. You can present two people – who believe malaria nets are equally effective – with some new data supporting the effectiveness of malaria nets, and they can arrive at different conclusions. Someone who’s more sceptical about some interventions being highly effective would be relatively unmoved. GiveWell have generally occupied this position. Someone relatively sceptical may decrease the tractability scores of most causes, except global health and some interventions within promoting effective altruism. Their list might start: promoting effective altruism, prioritisation research (still highly ranked due to importance and uncrowdedness), global health, immigration reform and ending factory farming.

A related issue is how much to trust common sense. If you think that generally people have developed somewhat effective ways to achieve their ends, then you’ll place higher weight on common sense. If on the other hand you think that common sense is often badly wrong and easy to beat without much extra research, then you’ll put less weight on it. We put moderate weight on common sense. Someone putting higher weight on it might start their list: global health, promoting effective altruism, ending factory farming, biomedical research, prioritisation research and developing world economic empowerment. Someone putting lower weight on it may start their list: prioritisation research, promoting effective altruism, global catastrophic risks, research policy and infrastructure, ending factory farming, improving decision-making and immigration reform.

Judgement calls about risk-aversion

Some people want to do good with a high level of certainty. We think, however, that high uncertainty doesn’t matter in itself. We believe that in principle, what you should do is weight each intervention by the good it would do, calculated by the probability of success (calculate the expected value). If one intervention yields 10 units of good with a probability of 10%, that’s just as good as an intervention which yields one unit of good with certainty. If you’re risk-averse about doing good, you’ll prefer the intervention that does good with certainty.

Someone more risk-averse than us, however, may disfavour global catastrophic risks, research causes and advocacy causes. Their list may start: promoting effective altruism, prioritisation research, global health, ending factory farming, and research policy and infrastructure.

Which causes are most robust under uncertainty?

In these alternative lists with different key judgement calls, the top couple of causes in the original list still score highly. That’s because just penalising them in one dimension, like importance, isn’t enough to make them unpromising overall, since they may still be high in the other dimensions. This gives us added confidence in the methodology, and is evidence for our earlier claim that generally prioritising causes is a useful project, despite differences in individual assumptions and values.

If someone differed from us on multiple judgement calls, however, then their list might become substantially different. For instance, someone who’s sceptical, less concerned with the long-run future, and places high weight on common sense may start their list: global health, biomedical research, developing world economic empowerment, promoting effective altruism and ending factory farming.

Our research process

The framework and general process for assessing causes used is explained here.

We applied this framework by:
1. Gathering promising causes from leaders in prioritisation research (especially GiveWell, the CEA and the Copenhagen Consensus), our general knowledge and major strategic foundations, like the Gates Foundation.
2. Intuitively applying the framework to these causes, with our basic reasoning for each cause explained above.
3. Discussing causes heavily with people from the CEA Strategy Research team, who in turn are in touch with the Future of Humanity Institute and other groups within effective altruism.

We should emphasise that this list involves many potentially controversial judgement calls, and we expect it to change significantly as more evidence comes in.

We plan to take this research forward by:
1. Continuing to update the list based on new findings from the CEA’s new prioritisation research team, GiveWell Labs, the Copenhagen Consensus and other new prioritisation research groups.
2. Further explaining and deepening our research into specific causes as they arise during case studies.

Some extra causes we considered including are: boosting technological progress, especially R&D to increase crop yields and green energy R&D (highly rated by the Copenhagen Consensus), basic science (which involves a substantial market failure), some sub-causes within education and governance (highly weighted by common sense, with potentially good long-run effects), political innovation, and trade reform (which, similarly to immigration reform, has the potential to have a huge positive impact on the global poor, and is highly rated by the Copenhagen Consensus).

Other questions

Why is this different from GiveWell’s list of top charities?

There are several reasons. First, GiveWell primarily aims to find the best funding opportunities, so it has a different perspective to us. The best funding opportunities are very relevant to people pursuing earning-to-give, but we expect them to be different from the best opportunities to deploy your human capital. Malaria nets are a good funding opportunity precisely because they require relatively little additional skilled human capital. All mainly required is more nets to be made and shipped, which by this stage can easily be accomplished with additional money. We’re much less sure that working on the malaria nets intervention is the best thing to do with your human capital.

We’re aiming to find the best causes for you to generally work within for at least the next couple of years. Our focus is broader than finding non-profits with funding gaps. For instance, there are many promising activities within research and government, and we don’t think it’s obvious that you shouldn’t work on them rather than find a non-profit to support. We’re also focusing on a smaller scale than GiveWell, which aims to find charities with room for at least US$1m more funding, whereas some of the causes on this list would struggle to absorb that many resources at comparable rates of return.

Second, we think that GiveWell is more effective than their top recommended charities (reasons in their own words), but they don’t recommend themselves for preserving impartiality. GiveWell is an example of prioritisation research and promoting effective altruism, and we think the effectiveness of GiveWell, among other organisations, is evidence that these causes are in general more promising than global health (at least on the scale of investing less than US$1m).

Third, we think we differ somewhat from GiveWell in our framework and key judgement calls. In particular, we think GiveWell might underweight the importance of long-run flow-through effects, animal welfare and value of information, while placing higher weight on common sense, and being more sceptical about the ease of finding unusually good interventions.

Don’t the answers depend on the person?

People can easily disagree over which cause is most generally effective, due to disagreements over key judgement calls – see our examples above.

It’s also important to clarify that “which cause is it best for you to support” is a separate question. That’s because different people have different types of human capital and other resources to contribute. Since some causes are more in need of some types of human capital than others, different people should support different causes.

Different people also have different comparative advantages and levels of motivation, which can also alter their choice of cause.

Consider using our list as a starting point, combined with other particularly promising opportunities you know of. Think through how you might differ over key judgement calls, and then work out where you can make the biggest contribution within that list.

We plan to write more about which causes require different types of skills as the issue comes up in future case studies.

How do you judge effectiveness?

This post explains our framework and process for assessing causes, in terms of effectiveness.

Is there a cause you think is good that hasn’t been included? Post a write-up of it below!


Thank you to Carl Shulman, Jonah Sinick and Nick Beckstead for comments, though they may not endorse all of the claims made.


You might also be interested in:


Notes and References


  1. Though there are also biases which cause overinvestment in some cases, e.g. investment in protection against terrorist attacks. 
  2. A written testimony to the House Committee on Science and Technology Hearing by Phil Rasch (2010).