Related: Purchase Fuzzies and Utilons Separately

We genuinely want to do good in the world; but also, we want to feel as if we're doing good, via heuristics that have been hammered into our brains over the course of our social evolution. The interaction between these impulses (in areas like scope insensitivity, refusal to quantify sacred values, etc.) can lead to massive diminution of charitable impact, and can also suck the fun out of the whole process. Even if it's much better to write a big check at the end of the year to the charity with the greatest expected impact than it is to take off work every Thursday afternoon and volunteer at the pet pound, it sure doesn't feel as rewarding. And of course, we're very good at finding excuses to stop doing costly things that don't feel rewarding, or at least to put them off.

But if there's one thing I've learned here, it's that lamenting our irrationality should wait until one's properly searched for a good hack. And I think I've found one.

Not just that, but I've tested it out for you already.

This summer, I had just gone through the usual experience of being asked for money for a nice but inefficient cause, turning them down, and feeling a bit bad about it. I made a mental note to donate some money to a more efficient cause, but worried that I'd forget about it; it's too much work to make a bunch of small donations over the year (plus, if done by credit card, the fees take a bigger cut that way) and there's no way I'd remember that day at the end of the year.

Unless, that is, I found some way to keep track of it.

So I made up several jars with the names of charities I found efficient (SIAI and VillageReach) and kept a bunch of poker chips near them. Starting then, whenever I felt like doing a good deed (and especially if I'd passed up an opportunity to do a less efficient one), I'd take a chip of an appropriate value and toss it in the jar of my choice. I have to say, this gave me much more in the way of warm fuzzies than if I'd just waited and made up a number at the end of the year.

And now I've added up and made my contributions: $1,370 to SIAI and $566 to VillageReach.

A couple of notes:

  • I do think it was a good idea in practice to diversify my portfolio (despite the usual admonitions to the contrary) because it appeared to increase my charity budget rather than divert a fixed one. Some days I just didn't feel as optimistic about the SIAI, and on those days I could still chip in to save lives in the Third World. As long as my different jars seem to be interfering constructively rather than destructively, I'll keep them.
  • In terms of warm fuzzies, I really enjoy that this system makes giving more tangible than writing a check or filling out an online form. It even helps that I have the weighted clay chips- tossing those into a jar feels as if I'm actually doing something.
  • I do worry about doing my good deed for the day and having negative externalities flow from that, so I do my donating at the end of the day to minimize the effect.
  • I could easily afford to give more than this, actually (though I can't tell whether I would have– it's more than I donated to charity in any previous year, although I was a poor grad student until this fall); I'm going to see if that knowledge makes me increase my pace of giving next year. (UPDATE 8/19/14: In retrospect, it was much more important for my less wealthy past self to create a habit than for him to donate a significant fraction of his income. My contributions to the chip jar since then have scaled appropriately to my circumstances.)

Let me know if you start trying this out, or if you have any suggested improvements on it. In any case, may your altruism be effective and full of fuzzies!

ADDED 12/26/13: I've continued to use this habit, and I still totally endorse it! A few addenda:


  • I've now labeled the jars "Maximally Effective Altruism" and "Directly Helping People Now", and I wait to decide where to direct each of those jars until I'm ready to make my donations.
  • One little fuzzy bonus: I find it pretty fulfilling throughout the year whenever I have to consolidate my lower-denomination chips into larger-denomination ones.
  • If you're new to the idea of effective altruism (aiming not simply to do good for the world, but to try and do the most good possible (in expected value) with your donation), this essay is an awesome introduction, and organizations like GiveWell and Giving What We Can exist to help make it easier.


New Comment
49 comments, sorted by Click to highlight new comments since:

Once my workplace had a party/fair allegedly to raise money for some charity.

I was slightly miffed to the low util to fuzzies ratio, and to the company's taking the credit for the employee's fundraising, with no corporate matching.

So, when I was asked for money at the event (one-on-one, not in front of everyone), I wrote a check to my favorite charity, for about the same total as the entire fundraiser, right in front of the person asking for the money; I explained myself politely and the requester (I think) took it as an impressive act of charity rather than as asociality. The check was in addition to my usual monthly donation.

Moved to LW main and promoted.

Let me know if this seems like a bad idea for some reason, but when something gets 34 upvotes...

I think it might be helpful to set a numerical cutoff and perhaps an editorial policy (EG so that terribly edited content or pure links don't hit the front page) and then stick to that policy. Before seeing what you did here, I briefly thought that this sort of action would be good for the recent vegetarianism discussion, simply because it had more than 20 upvotes and raised apparently legitimate issues about socially calibrating oneself on pragmatic moral issues that require abstract thinking.

Of course, that discussion implicitly criticized SIAI as visible non-vegetarians. The implicit potential danger of promoting this but not that is that this can be seen as consistent with promoting things that build SIAI and LW organizationally and thus clearly benefits yourself as the founder. Some readers might infer the existence of such a policy and think it is off mission relative to the "refining the fine art of human rationality" tagline.

The lack of one intervention can give lie to an explanation for an second intervention if the explanation would justify things that weren't done, like the way North Korea clearly had WMDs (and no oil reserves) but wasn't invaded, but the details of why it "didn't count" were never spelled out in plain english by policy makers, leaving people free to speculate.

Honestly, I think organizational development is instrumentally critical to the tagline mission, because institutions and F2F stuff can do a lot more a lot faster than mere blog posts, and there is all kinds of good literature to this effect. But these sorts of issues can be very tricky to get right if the inferential distance between the readership and the mods gets too large. The connection between institutional growth and front line effort isn't always obvious to everyone.

Having a bright line editorial policy that can be seen to promote content that doesn't obviously build the organization is probably useful for visibly signaling good faith from above. Another approach might be to directly explain part of the rational basis for pursuing certain kinds of institutional growth, so that instead of justifying this with "34 votes" you could have justified this with "34 votes and efficaciously pro-social".

(Also, on a general note, the front page is pretty awesome right now, with more solid content and less meetup stuff than has been normal for a while. If someone is doing something to consciously bring about this state of affairs, they deserve credit.)

Thanks! I'm fine with that.

I'd have tried to polish it up more if I expected it to get promoted, but in that case I might have put off writing it altogether.

The post's upvote score now shows as zero (a dot), even when I add/remove my own vote.

Dot doesn't indicate 0, dot indicates "this is new so we're not displaying a score".

Ah! Thanks for the explanation. And indeed now it's displaying a score.

Are you using the old link to the post in the Discussion section or accessing it from the main page?

From the main page.

Small note: the post originally had my second charity as the Stop TB Partnership (GiveWell's second-place charity) rather than VillageReach, essentially on the theory that if everyone on GiveWell only donates to the very top charity, then other charities have no incentive to become more transparent unless they can claim the top spot.

Then I went to actually make my donation, and my warm fuzzies were interrupted by the donation process. I switched back to VillageReach, whose donations are handled much more efficiently.

Yeah, this is one of those minor issues, but I think it's really important for my future willingness to donate that I have a good first experience and no nagging doubts about the process.

Just came up with a nice idea for a very good first experience:

Put actual money in the Jar. When the year ends, make the donation through Check or Online.

Match the donation you just made with a donation for yourself, an yearly gift that your previous altruist selves gave you to spend in happiness increasing activities. That is, right after finishing your donation, open the Jar, put the money in your pocket, and start thinking about how to invest it in happiness.

Suggest turning this into LW main post. (You can Edit and re-save it there.)

I would like to know SIAI's official position on the Slate article and its suggestions to prospective donors.

Landsburg is correct about what rational agents should do. (Period.)

Human altruists may have to resort to more complex tactics like

I don't see why SIAI should have an official position on this.


I don't see why SIAI should have an official position on this.

Excellent hack!

If it turns out there is a good-deed-for-the-day effect in this context, perhaps one could use another set of jars for purchasing anti-fuzzies with selfish utility. For example if you had one for payments into a cryonics annuity, you might then feel compelled to put equal or greater amounts into the charitable jars to balance it out.

Or better yet: books and good beer. The "selfish jar" needs to be something that pays out now or in the near-term.

I've been vaguely thinking that someone should make an electronic version of this for a while (I think I might have seen it suggested somewhere else on LW as well?) - a Givewell iphone app, so that when someone on the street asks you for money, you can immediately, there and then, donate it to a more worthy cause, while you're still feeling guilty about not giving money to the homeless person who was only going to spend it on alcohol anyway.

I doubt this would be too much work for someone with any experience, but I have no knowledge of writing iphone Apps and no Mac - is there someone out there with the means and the will to actually do it?

I like it-- this seems like a good approximation of the "instant donate-a-dollar button" that Marcello (I think) suggested as a potential iPhone app earlier. Does anyone know if progress has been made on such an app, by the way?

My father spent some time creating such an app, but I don't think it is useful for the purpose of diverting sudden, altruistic impulses toward high impact charities, largely because it requires you to manually input your credit card/paypal info each time you try a donation, which is enough effort to reintroduce the original trivial inconvenience. If anyone knows how to fix that sort of problem, I could get the code for you.

I don't have to input my financial info when I use paypal - I just log in with my username and password, which can be cached for those who don't like even that inconvenience; and click on "Pay now".

Yeah, I think the point was that the app would have some way to "preload" that information so you wouldn't have to reenter it every time. I'm not sure if that's viable from a coding perspective, but it seems like it would work to solve the main problem here.

I think this is an excellent idea. Members of certain religious traditions do something similar, by having literal charity boxes they keep at home; this allows them to donate at the moment rather than allow the moment to pass. Yours obviously updates the idea: people keep much less of their money in physical form, so the change to chips may be very helpful.

I wanted to comment on "I do worry about doing my good deed for the day and having negative externalities flow from that, but I can't say I've seen it happening yet." I think this is always a real problem, and I have a few possible suggestions.

  1. When possible, try to donate later in the day rather than earlier. (but of course, you don't want to let the inspiration pass, so don't take this too far).
  2. When you do donate earlier, you can then remind yourself of the early donation, and therefore the need to find ways to be helpful to others. Alternatively, lsparrish and you talk about the antifuzzy jar for beer/etc. Perhaps the ties should go both ways; by rewarding yourself explicitly for your good behavior you may potentially see that good deed as "resolved", and avoid trying to compensate by mistreating others. I don't have data on that, but it would be interesting to look at (I may try this myself over the next few months).
  3. Make sure the chips only substitute for financial donations. If you let it substitute for helping people with your time/effort, there may be less positive consequences.

Why not just make an automated monthly payment (through PayPal, your bank, or your credit card)? For tangibility (but more temptation to slip), write and snailmail a check each month.

Why not just make an automated monthly payment (through PayPal, your bank, or your credit card)?

Because that is nearly the opposite of the kind of experience he was looking for.

What he said.