You are currently browsing the tag archive for the ‘effective altruism’ tag.
The power of exponential growth seems to make a compelling case for effective altruists to delay their donations. An average 5% return on investment (ROI) would turn one dollar into ten in 50 years time. If saving a life costs $2000 now and similar opportunities will exist in the future it would cost just $200 to save a life in 2062 – a relative bargain! Sadly things aren’t so simple. Whether we really should delay depends on specifics of the activities we are funding and difficult predictions about the future. Here I’ll summarise the most important uncertainties as a roadmap for future posts.
Our goal can be summarised as choosing the time t which maximises
(1 + Return on investment)t × Cost effectiveness of donationt
× Probability of donation actually being madet.
Unless you are a multimillionaire, the relevant expected ROI is the highest one available without regard to risk. Giving $2m will do about twice as much good for the world as $1m, so to maximise your expected impact you should just maximise your expected donation. Note that if your favourite charity would be able to use money now to attract donations at a rate faster than you expect your investments could return profits then donating would have to be better.
The second and more challenging issue is how cost effective your donation will be in the future relative to now. If you thought basic health would be the optimal cause this would involve anticipating things like
- the extent of poverty
- the cost of delivering health services
- how much other donors will be funding the low hanging fruit.
The last point is especially relevant for those like me thinking of funding existential risk reduction because a few billion from governments or philanthropists could make a big impact on the value of further funding in that area.
In evaluating cost effectiveness we must factor in that any good charity will have impacts that propagate through time and so offer its own ROI. For instance, combatting contagious diseases now rather than in 2062 should lead to fewer people becoming infected in the meantime and so result in a richer and healthier population in 2062. Similarly, spending on existential risk reduction draws attention, money and researchers to that issue. Giving now leaves your donations more time to have this snowball effect during the window of greatest extinction risk.
On the other hand delaying leaves you more time to identify cost-effective targets for donations. Personally, I am investing rather than giving mostly because I expect groups like 80,000 Hours to give me a much better idea of how to best reduce existential risk within the next decade.
Finally you must assess the risk of your donation never being made, for example due to a catastrophe which eliminates your savings. If you can’t bind yourself through a trust fund, you must also worry about changes to you or your life which result in you deciding not to give.
Last year I gave a few thousand dollars to the charity Village Reach, which performs vaccinations in rural Mozambique, on the recommendation of charity evaluator GiveWell. The bottom line of the recommendation is that Village Reach can prevent a childhood death for around $400-$800 – very cheap indeed.
A natural response might be to ask what value there is in saving someone’s life in such a poor country. Though I don’t know a great deal about quality of life in rural Mozambique, living in a country with “one of the lowest GDP per capita, one of the worst human development index and amongst the highest inequality in the world” presumably isn’t great. Anywhere that you can save lives for a few hundred dollars would have to be pretty rough! Given this, the justification for vaccinations has to rest on more than just averting death. It must also be about improving people’s quality of life.
Population effects aside, reducing the rate at which people die is a significant way to improve quality of life. Being sick and dying is painful for the person involved and those who care for them. Further, raising kids only to have them die in infancy uses up resources in a community that has no resources to waste. Childhood disease reduces the intelligence and health of survivors and depresses school attendance. A high risk of a child dying discourages investment in human capital and encourages large families, both of which are probably bad for economic development.
If we want to do as much good as possible with health interventions we should aim to not only avert direct suffering from disease and death. The treatments that will most effectively improve quality of life in the long run will also spur on the economic development that allows people to support themselves.
What’s more, for someone who wants to maximise ‘total welfare’, the impact health spending has on population is not a second-tier issue. If Village Reach improves health without increasing incomes or reducing fertility, then it may just result in more people living in abject poverty, which is a questionable achievement. On the other hand folks who are optimistic about the quality of life of people living in poverty will not be so enthusiastic about fertility falling unless the population decline does a lot to improve average quality of life.
It is much harder to quantify these flow on effects on development and population, which is why they usually get short shrift. Education, health and development all cause one another with different intensities and lags, and unravelling the chains of causation between them is extremely difficult, if not impossible, in the absence of randomised experiments. A charity like Village Reach could randomly allocate villages to treatment and non-treatment groups and collect data on incomes and fertility in addition to health. Tracking these effects would require collecting data for an extended period of time, but would be a very valuable research project in its own right.
GiveWell has chosen to focus on health and nutrition interventions over others in large part because many more of them are cheap and have proven impacts.  Strong evidence of cost effectiveness is key for GiveWell, as a large part of what they are trying to achieve is a shift the culture among NGOs towards thorough data collection and evaluation of projects.  Given the current low standards of evaluation for most charities, this is a creditable goal.
A manageable improvement given this constraint would be to look at which kinds of disease do the most to depress education and productivity. A long lasting tropical parasite, childhood diarrhoea, chronic illness or fatal adult disease could all have different impacts on family structure and capacity for education and work. Likewise some countries may be in a better position to advance economic development in response to improved health than others. These flow on effects may be as important, if not more so, than the number of deaths averted per dollar.
 An education program which can’t demonstrate an impact on education presumably isn’t doing much for quality of life, population or development either, so it is fair enough to ignore it.
 While GiveWell’s evaluation style is likely to be biased towards interventions that have easily measured, short-term outcomes, this isn’t a problem necessarily. While GiveWell may miss highly effective charities, something which can’t be measured can’t be targetted.
People often say to me that certain actions, outcomes or policies are ‘good’. Something I usually try to establish right away is ‘compared to what?’ There are three common responses: relative to nothing, relative to the status quo and relative to anything else. In most circumstances the first two comparisons are red herrings; the standard we should usually instinctively apply is the last.
In personal decisions this is clear enough. If you are trying to decide what to do and are going through programs you could watch on television, there is no use comparing each program to staring at a blank TV screen, to whatever channel it happens to be tuned to now, or to sitting and doing nothing. The fact that a show is better than a blank screen is irrelevant to your decision. Instead you should compare watching each program to the best alternative activity you can think of and only watch it if it is better.
When someone says that giving money to a charity is ‘a good thing to do,’ for some reason that standard tends to be relaxed. They ought to compare that charity to the best way to spend that money, not just to destroying it or spending it on themselves. For some reason wastefully spending the money on oneself is the standard comparison in this case. Similarly, when we consider a government policy, we should compare it to the best policy in that area that we can think of, not to doing nothing at all or continuing to do whatever we are already doing.
That’s not to say you should never consider other more limited comparisons. For example, we might restrict our comparison to ‘other charitable acts we would actually be willing to do’, or ‘other policies that would be able to get through parliament.’ If someone else thinks that doing nothing is the best option, then in a conversation with them we may want to use that as the comparison for the sake of argument. However, when we want to make a more limited claim we should ensure everyone understands what comparison we are drawing and why we have chosen that counterfactual.
I’ve heard a lot of conversations over the last few years about the merit or lack thereof of the Australian Government’s National Broadband Network rollout. Many people evaluate the network compared to ‘changing nothing’. Most others evaluate it relative to ‘the best alternative broadband policy I can think of,’ or ‘the best way to spend that money.’  Needless to say, the people who draw the first comparison tend to judge the NBN positively, while those who apply the latter standards are more likely to evaluate it negatively.  To my mind the first comparison would only be interesting if doing nothing were the likely alternative, which I’m pretty sure is not the case. We should at least ask whether the NBN is ‘good’ compared to other broadband policies, and ideally whether it is good relative to all the other things the government could spend that money on.
 I am taking no position on the merits of the NBN.
 If we are spending money to achieve some goal, at a minimum we should compare that expenditure to the best alternative way of achieving that goal we can find. This is called cost-effectiveness. Ideally we would go further and compare spending money on that goal to the best way we could spend money to achieve any goal. That is usually much more difficult to do, and requires agreement not only about whether that goal is desirable, but how important all goals are relative to one another. So if we going to dedicate some resources to achieving that original goal, it can be useful to search for the most cost effective way to do so without comparing it to all possible alternatives.
There is a principle in finance that obvious and guaranteed ways to make a lot of money, so called ‘arbitrages’, should not exist. It has a simple rationale. If market prices made it possible to trade assets around and in the process make a guaranteed profit, people would do it, in so doing shifting some prices up and others down. They would only stop making these trades once the prices had adjusted and the opportunity to make money had disappeared. While opportunities to make ‘free money’ appear all the time, they are quickly noticed and the behaviour of traders eliminates them. The logic of selfishness and competition mean the only remaining ways to make big money should involve risk taking, luck and hard work. This is the ’no arbitrage‘ principle.
Should a similar principle exist for selfless as well as selfish finance? When a guaranteed opportunity to do a lot of good for the world appears, philanthropists should notice and pounce on it, and only stop shifting resources into that activity once the opportunity has been exhausted. This wouldn’t work as quickly as arbitrage on financial markets of course. Rather it would look more like entrepreneurs searching for and exploiting opportunities to open new and profitable businesses. Still, in general competition to do good should make it challenging for an altruistic start-up or budding young philanthropist to beat existing charities at their own game.
There is a very important difference though. Most investors are looking to make money and to them a dollar is a dollar, whatever business activity it comes from. Competition between investors makes opportunities to get those dollars hard to find. The same is not true of altruists, who have very diverse preferences about who is most deserving of help and how we should help them; a ‘util’ from one charitable activity is not the same as a ‘util’ from another. This suggests that unlike in finance, we may able to find ‘altruistic arbitrages’, that is to say ‘opportunities to do a lot of good for the world that others have left unexploited.’
The rule is simple: target groups you care about that other people mostly don’t, and take advantage of strategies other people are biased against using. The rule is the root of a lot of advice offered to thoughtful givers and consequentialist-oriented folks. An obvious example of the rule is that you shouldn’t look to help poor people in rich countries. There are already a lot of government and private dollars chasing opportunities to assist them, so the low hanging fruit has all been used up and then some. The better value opportunities are going to be in poor, unromantic places you have never heard of, where fewer competing philanthropist dollars are directed. Similarly, you should think about taking high risk-high return strategies. Most do-gooders are searching for guaranteed and respectable opportunities to do a bit of good, rather than peculiar long-shot opportunities to do a lot of good. If you only care about the ‘expected‘ return to your charity, then you can do more by taking advantage of the quirky, improbable bets neglected by others.
Who do I personally care about more than others? For me the main candidates are animals, especially wild ones, and people who don’t yet exist and may never exist – interest groups that go largely ignored by the majority of humanity. What are the risky strategies I can employ to help these groups? Working on future technologies most people think are farcical naturally jumps to mind but I’m sure there are others and would love to hear them.
This principle is the main reason I am skeptical of mainstream political activism as a way to improve the world. If you are part of a significant worldwide movement, it’s unlikely that you’re working in a neglected area and exploiting how your altruistic preferences are distinct from those of others.
What other conclusions can we draw thinking about philanthropy in this way?
I’m going to do a back-of-the-envelope calculation on ‘local food’ and then later ‘fair trade’ to explain why I don’t think they are worth putting much effort into. I hope it will inspire you to do the same for whatever approaches you currently use to make the world a better place.
As you probably know, local food or locovorism is where people aspire to buy food made near to where they live. On a recent trip to the States I was astonished at the size of the local food movement. Everywhere I went people would beam with pride about how nearby their food was produced. There are several proposed benefits of local food  but probably the most common one is wanting to reduce the environmental damage caused by food transport, so called ‘food miles’. For this to be the best thing for someone to dedicate their attention to a few things need to be true:
- environmental degradation and climate change would need to be valuable things to direct attention to on the margin;
- reducing your personal environmental footprint would need to be an effective thing to do about those problems;
- buying food produced near you would need to be a cost-effective or effort-effective way to achieve that.
Let’s assume for the moment that environmental degradation were the most important problem to work on. Reducing the effect of your personal consumption would be only one way to direct your effort. Others would include working to change environmental regulations, convincing others to do more themselves, expanding humanity’s ability to adapt to environmental degradation (for instance through poverty alleviation, migration or geoengineering) or assisting researchers developing green technology. Intuitively I expect all of those to pack a bigger punch per hour than trying to change your own consumption habits. But let’s say my instinct is wrong about that.
If you wanted to reduce the environmental impact of your own consumption, would buying food produced nearby be an effective approach? Let’s get some indication of the good you could hope to do:
Desrochers and Shimizu cite a comprehensive study done by the United Kingdom’s Department of Environment, Food and Rural Affairs (DEFRA) which reported that 82 percent of food miles were generated within the U.K. Consumer shopping trips accounted for 48 percent and trucking for 31 percent of British food miles. Air freight amounted to less than 1 percent of food miles. In total, food transportation accounted for only 1.8 percent of Britain’s carbon dioxide emissions.
In the United States, a 2007 analysis found that transporting food from producers to retailers accounted for only 4 percent of greenhouse emissions related to food. According to a 2000 study, agriculture was responsible for 7.7 percent of total U.S. greenhouse gas emissions. In that study, food transport accounted for 14 percent of the greenhouse gas emissions associated with agriculture, which means that food transport is responsible for about 1 percent of total U.S. greenhouse gas emissions.
So if we assume that buying local food eliminated all emissions from food transport you could hope to cut 1-2% of your total greenhouse gas emissions. Then there are some offsetting effects. A strong preference among consumers for local food would tend to drive agriculture towards places where it is otherwise less efficient, requiring more machinery, labour or land to produce the same food. This would also be bad for the environment. On top of this, as indicated in the quote above, retail-to-home transport has about the same impact as farm-to-retail transport. If someone drives further to the farmers’ market to buy local, they could end up producing more food miles overall. How significant these offsetting effects are will vary depending on the person and the food they are buying, but they suggest that 2 per cent is a generous upper bound.
What would be the rough value of a 2 per cent reduction in your emissions? Let’s say you are a big polluter and produce 20 tonnes of CO2 equivalent each year. Let’s also take a high value of emissions reductions of $100 per tonne. This is several times the current marginal cost of reducing emissions and is probably more than enough to get us climate stabilisation at 450 CO2e, but let’s use it anyway. A 2% reduction in your emissions would then be worth $40 of good to the world (20 tonnes * $100 * 2%). This seems small to me for a year’s work buying local food, and that is the best case scenario. Insulating your home, not eating livestock or holidaying by plane, or buying up and ‘retiring’ carbon credits in carbon markets are likely to offer much more bang for buck.
Could eating local still be worthwhile? Sure. If you would eat food produced nearby for selfish rather than altruistic reasons, or are tossing up between a local and foreign item it’s little or no difference to you to choose the local one, go ahead. But if your goal is to effectively convert your time and money into good outcomes for the world it would be very surprising if ‘eating local’ were something worth making a fuss about.
 Other reasons I’ve heard offered are ‘I enjoy having a relationship with the farmer/land or helping small growers’ or ‘I like knowing how the animals/land are treated’. Insofar as you are eating local food because you enjoy it more you can disregard this post. As for being more informed about the effects of the farming techniques employed, or wanting to support small scale farming over big the same considerations apply. Does this really offer a high return on your effort? My other question would be: should you really have to eat locally to know how your food is produced? It seems like a less elegant solution than certification labels like ‘organic’ or ‘free range’.