Duke has a culture of philanthropy—and that’s awesome. You can’t walk across the Bryan Center Plaza without being flagged by numerous student groups fundraising for various charities. And the closer I get to graduating, the more subliminal messages I get from Duke trying to pressure me into giving back to the university. #ForeverDuke! Assuming I’m lucky enough to have money to give back, how do I decide where my money should go? At the end of the day, I have a finite amount of money I can donate, so there are obvious tradeoffs: a dollar I give to Charity A is one less dollar I could give to Charity B. And I’m only talking about this in context of donation, but I face a similar problem if I want to have a large social impact with my career.
In case you thought this was easy, I’ll try to make it harder for you—ask yourself, how do you decide between working at or donating to the American Civil Liberties Union, The Jameel Poverty Action Lab, Teach for America, Duke or the Against Malaria Foundation? Most people recognize the need to make sacrifices for the long term, hence the importance of solving climate change—how do we weigh the considerations of people now vs. people in the future? Is overpopulation something we should be worried about? What’s worse—living in extreme poverty or having malaria? Politicians have a lot of influence, maybe that’s a good way to make change? Unfortunately, discerning where we can have maximal impact is intractable, so we use approximations.
To compare problems, we need to estimate the amount of good done per extra person or dollar invested in each problem. We can break this down into three components: scale (amount good done / percent problem solved), tractability (percent problem solved / percent increase in resources), and neglectedness (percent increase in resources / extra person or dollar). The nice thing about this framework is that we can estimate each of these criteria separately, then multiply them to get back amount good done / extra person or dollar. (This framework was originally developed at the Open Philanthropy Project, then expanded upon by 80,000 Hours and the Future of Humanity Institute.)
Why do we want to put numbers on things as opposed to just use intuition to make decisions? I can feel empathy for a person impacted by a tragedy, but my brain literally cannot even begin to comprehend the suffering of almost a billion people living in extreme poverty. This idea is called scope insensitivity. It’s no wonder that many people don’t care about big problems as much as they should—they literally can’t. The point is not to make these estimates and follow them blindly, but to correct for the ways in which our intuition about big numbers fails us.
Scale (amount good done / percent problem solved)
If this problem was solved, how much better would the world be? This requires that we have some metric for impact. If we’re looking at global health, a standard way to measure wellbeing is by a unit called quality-adjusted life years (QALY), where one unit is equivalent to one year of healthy life. The idea of a QALY is to capture our intuition that intensity and duration both matter when considering harm done (i.e. one harm is equivalent to another harm that is twice as bad, but experienced for half the amount of time.) In terms of QALYs, cancer is responsible for 8 percent of all ill health world-wide, compared with 2.7 percent for malaria. Often we can get fairly reliable estimates of this when it comes to disease, whereas it can be unclear how much harm is caused to people by living under a non-democratic regime. If this sounds difficult, that’s because it is! But when you get a lot of smart people working on something, you make progress.
Tractability (percent problem solved / percent increase in resources)
Problems are only worth working on if we can actually find a solution. Furthermore, all else held equal, we should tackle easier problems before hard problems. The question of identifying which problems are easiest to solve is difficult, because it depends on assessing the effectiveness of our best solutions to the problem. If no good organizations exist, the question becomes, how hard would it be to establish an organization that could make a large impact in this area? How difficult is it to institute land reform policies in developing countries? Whether or not land reform is solvable or impossible could determine whether this is at the top or bottom of our list, and so it merits serious investigation.
Neglectedness (percent increase in resources / extra person or $)
We want to look for areas to help that are not “crowded.” If there are a lot of people already working on some problem, then often they will have already found all the low-hanging fruit. On the other hand, problems without much support, but of equal importance, benefit more from every extra person or dollar—this claim is the well-known law of diminishing returns.
In sum, we want to find large, solvable problems that not many people are working on. You might be wondering: Effective Altruism did not invent the idea of helping people efficiently, so why should we think that we should be able to do any better than anyone else has done in the past? Some problems that meet our outlined criteria already exist in the public consciousness, like global poverty and climate change. However, this framework also suggests that there are high-priority problems that are seriously undervalued: risks from emerging technologies, factory farming, improving institutional decision making, and more promising areas. The key is that you have to look where no one else is looking. As a result, some results may be counterintuitive, and this is by design.
In subsequent columns, we’ll explore some problems that seem especially urgent, and how you can you use your donations and career to help solve them.
Get The Chronicle straight to your inbox
Signup for our editorially curated, weekly newsletter. Cancel at any time.
This column was written by Miles Turpin, a Trinity junior, with input from members of Effective Altruism: Duke. If you have ideas, criticisms, or questions regarding Effective Altruism, reach out via firstname.lastname@example.org.