{"items": [{"author": "Neela", "source_link": "https://www.facebook.com/jefftk/posts/819471645132?comment_id=819473162092", "anchor": "fb-819473162092", "service": "fb", "text": "Interesting. I've actually never thought the donors would send the amount anyway - I thought it was a donation conditional on raising the funds, a true match, so no match if the target isn't met. Eye-opening! I suppose, though that is doubling the impact of the donor's money if you match what they would have donated and you are an incremental donor.", "timestamp": "1476994070"}, {"author": "Jeff Kaufman", "source_link": "https://www.facebook.com/jefftk/posts/819471645132?comment_id=819473162092&reply_comment_id=819473526362", "anchor": "fb-819473162092_819473526362", "service": "fb", "text": "→ What's especially bad is when a charity uses your argument when talking to the sponsor to show how their money is going to have a 2x impact via incremental donors, and then turns around and tells the incremental donors that they should donate now to take advantage of the 2x match and double their impact.", "timestamp": "1476994306"}, {"author": "David", "source_link": "https://www.facebook.com/jefftk/posts/819471645132?comment_id=819475512382", "anchor": "fb-819475512382", "service": "fb", "text": "\"Other EA and EA-aligned organizations have run matching campaigns, however, because offering matching funds does bring in more donations.\"
What's the evidence for that? It seems easy to demonstrate that matching would cause an increase over some baseline while the matching is offered. But how do they demonstrate that matching has an effect on the total donations over time? (Anecdote: I chose to donate to GiveWell in part because their writing about things like matching convinced me that they have a nuanced understanding of reality, and are unlikely to try to fool donors.)", "timestamp": "1476995365"}, {"author": "Jeff Kaufman", "source_link": "https://www.facebook.com/jefftk/posts/819471645132?comment_id=819475512382&reply_comment_id=819475826752", "anchor": "fb-819475512382_819475826752", "service": "fb", "text": "→ I think there's pretty good evidence on matching helping bring in more money based on things like A/B testing direct mail charity solicitations. If you do this for charities people wouldn't otherwise have donated to at all, then the increase you see with matching is real and not just people adjusting when they're making their donations.", "timestamp": "1476995555"}, {"author": "David", "source_link": "https://www.facebook.com/jefftk/posts/819471645132?comment_id=819475512382&reply_comment_id=819476719962", "anchor": "fb-819475512382_819476719962", "service": "fb", "text": "→ That would just indicate that matching causes an increase from people who wouldn't have donated otherwise. I was speculating that matching might also cause a decrease from people who would have donated if there had been no matching. I don't think A/B testing at the donor level would help here, unless you can prevent a potential donor from finding out about matching campaigns.", "timestamp": "1476996170"}, {"author": "Ben", "source_link": "https://www.facebook.com/jefftk/posts/819471645132?comment_id=819475512382&reply_comment_id=819518466302", "anchor": "fb-819475512382_819518466302", "service": "fb", "text": "→ Jeff Kaufman do you have links on evidence you mentioned via A/B testing of direct mail campaigns (if they're ones that don't appear in http://www.benkuhn.net/matching )?", "timestamp": "1477015650"}, {"author": "Jeff Kaufman", "source_link": "https://www.facebook.com/jefftk/posts/819471645132?comment_id=819475512382&reply_comment_id=819600786332", "anchor": "fb-819475512382_819600786332", "service": "fb", "text": "→ @David: Ignore me, read Ben's link.", "timestamp": "1477056461"}, {"author": "Neela", "source_link": "https://www.facebook.com/jefftk/posts/819471645132?comment_id=819485726912", "anchor": "fb-819485726912", "service": "fb", "text": "But if the organization says \"someone will match up to xx\" do you really think the matcher donates even though the pledge wasn't met?", "timestamp": "1476998738"}, {"author": "Jan-Willem", "source_link": "https://plus.google.com/100580955183019057735", "anchor": "gp-1477004611802", "service": "gp", "text": "Agree on both counts. I'll also note that in the office-directed \"matching\" campaigns it's also been a great way to raise attention for EA in general and for particular effective charities. But that's because third parties who don't necessarily know about the charities are being told about the match, and it's being offered by folks that they \ndo\n know along with a blurb about why giving to that charity makes sense.", "timestamp": 1477004611}, {"author": "Elizabeth", "source_link": "https://www.facebook.com/jefftk/posts/819471645132?comment_id=819521165892", "anchor": "fb-819521165892", "service": "fb", "text": "I found GiveDirectly -> GiveWell -> EA through the google employee matching program, so I am pretty well disposed to it", "timestamp": "1477017135"}, {"author": "Benjamin", "source_link": "https://www.facebook.com/jefftk/posts/819471645132?comment_id=820815092852", "anchor": "fb-820815092852", "service": "fb", "text": "80k has run donation matches but we've tried to make them partially counterfactually valid. Usually the matching donor says that if the funds aren't matched, they won't donate to 80k for at least a year. Of course that means we may get the funds eventually, but waiting a year is a significant cost and it's not certain.", "timestamp": "1477639552"}]}