Giving what we will

The effective altruism (EA) movement is still small, but I see what must be a disproportionate amount of discussion about it. Some of this comes from friends on Facebook, some comes from Scott Alexander at Slate Star Codex, and some comes most recently from Ozy Frantz at Thing of Things.

For those who don’t know about EA, the first few sentences of its Wikipedia article will be a good enough introduction: “Effective altruism is a philosophy and social movement which applies evidence and reason to determining the most effective ways to improve the world. Effective altruists consider all causes and actions, and then act in the way that brings about the greatest positive impact. It is this broad evidence-based approach that distinguishes effective altruism from traditional altruism or charity.”

Some of the discussion I see is about straightforward in-group concerns: which charities do the most good per dollar, how can we compare investments that are high-risk-high-reward against those that are low-risk-low-reward, etc. A remarkable amount of discussion, on the other hand, is about spreading the word. How do we get more people interested in charity? How do we get people who already donate to charity to exercise more discretion about what their donations achieve? How can we tell people that we give 10% or more of our income to charities without them getting defensive? And so on.

These latter questions are very important. One of the EA hubs is Giving What We Can (GWWC), which encourages people to pledge 10% of their income toward effective charities. Each additional pledge is a huge win in terms of human impact. GWWC claims that 10% of my annual (graduate student!) income would be expected to save a human life in the hands of a high-impact charity. They currently have only 670 members.

This post serves a couple of distinct purposes: first, it describes two new proposals that might get more people to donate to charity; second, it’s an announcement of my own intention to follow one of those proposals.

The GWWC Model

Presumably, GWWC uses 10% because it’s a round number, in some cultural contexts it’s already associated with the tradition of tithing, and it’s enough to make an impact without being so much as to scare every last person away.

For me, it’s too abstract. When I think about 10%, I can’t help but actually do the math. A big annual income divided by 10 is a pretty big more-than-monthly income.

That’s good, in one sense: goodness is proportional to bigness! But it’s not the right image to have in my mind when convincing me to do it.

Proposal 1: Divide and conquer

It would help me if the act of giving were framed in terms of something concrete that I actually did, rather than a proportion of my total income. In principle, what I’m doing isn’t the point; the outcome is what matters. But getting my monkey brain to share some of its tens of thousands of bananas is a different matter entirely.

To remove a layer of abstraction, I think about donations in terms of labor. Any donation that amounts to a modest 0.4% of annual income is equivalent to an entire day of work, assuming a five-day work week and two weeks of vacation per year. Donating 5% of income is equivalent to working for charity for one day each month.

Compare this to donating a day’s time each month directly to volunteer work. For the most part, volunteer jobs are highly unspecialized, and the labor I could do there is not worth as much as the equivalent labor in my day job, which requires stricter qualifications and higher levels of training. Further, my off-work time is precious to me, and it is psychologically easier to donate the labor I am already engaged in than to donate labor in the time that I normally reserve for taking care of myself.

There is one more symbolic advantage to thinking about donations in terms of a day’s labor each month. I can choose to donate my day’s labor on the last work day each month; my needs come first, but there is room left over for helping others. In a typical month, this comes out to working to support myself for almost 20 days each month and taking time off during weekends or holidays to take care of myself in other ways for almost 10 days each month, with just one day at the end when I’m working to give. That’s a schedule I can manage.

An additional and accidental advantage of earning to give at the end of each month is that this lines up well with Thanksgiving and Christmas, when people are most likely to be thinking about charity anyway.

Maybe most importantly, the resultant 5% of my income is psychologically and practically much less difficult for me to part with than 10%.

Thinking about charity in terms of days’ labor instead of percentage of income can obviously be extended beyond the specific one-day-per-month approach I’m proposing here. If you want to donate 10% of your income, then you can think of this as donating two days’ labor each month. If you’re really strapped for cash one month, then you can forego donations knowing that you’re working to take care of yourself in the time that you might otherwise be working for others, and this might avoid some of the guilt that might be entailed by falling short of your target percentage.

Last year, my New Year’s Resolution was to finish things that I started. This blog is partly an effort to facilitate that goal, and it’s gone well on the whole. This year, my resolution (just a couple of months early) is to earn to give for one day at the end of each month.

Proposal 2: Get them while they’re young

An alternative way to foster charitable giving, closer in spirit to the approach advocated by GWWC, is encouraging people to pledge an increasing proportion of their income starting from a relatively young age. Specifically, you could ask 21-year-olds to donate 1% of their income, with an additional percentage point each year until they’re donating 10% from the age of 30.

This would have to begin as a movement on college campuses. A lot of college students would be happy to make this sort of pledge because it wouldn’t cost them very much. When I was 21, I probably made about $5000, and 1% of $5000 is fifty bucks.

Another advantage of this approach is that young people are mostly poor, so it’s harder for them to give away the same proportion of their income as older, richer people. Giving an increasing proportion of money toward charity crudely accounts for this while setting an accessible benchmark so that people will still give something rather than nothing.

Most importantly, it’s much easier to recalibrate your expenditures when you’re only adjusting them by a small amount—say, from 0% to 1% of total income, or from 9% to 10%. Once people have made the pledge, they can work their way up to a high level of donations without leaving their comfort zones.

As with everything sociological, this would work best as a ritual. Perhaps each person could start donating on their birthday. Giving back to society in a structured, voluntary way would at the very least be a more meaningful symbolic entree into adulthood than getting really, really drunk (not that the two are mutually exclusive).


The best rituals, however, aren’t planned as a piece from the outset. I believe that as social technology evolves, structures to support charitable giving will take root in our culture in ways that I can’t yet imagine. The emergence of the EA community is a step in this direction. Their disciplined study of what works and what doesn’t leaves them in a promising position to push a few steps further.

For my part, at least, the biggest hurdle to giving was conceptualizing it in a concrete and nonthreatening way. Reducing a leap of faith to just another step is a huge step in itself. New models of structured giving need to be developed to expand the group of people who are comfortable giving in the first place. The GWWC pledge hasn’t worked for me, but fortunately it works for other people. Monthly “earn to give” days work for me but won’t for everyone.

There are still many possibilities to explore. In the meantime, those interested in giving might like to see the recommendations of GiveWell. For evaluations of organizations spanning a broader scope, CharityWatch makes their letter-grade ratings of many charities public, though more detailed evaluations require a donation to offset their research expenses.

Leave a Reply

Your email address will not be published.

To write LaTeX output, type $latex [input]$ with [input] replaced by LaTeX input.