Part of your critique is mostly valid in cases where donors have a fixed donation budget and allocate it to the best cause they come across, taking into account a potential leverage factor. I wonder whether instead a lot of the donors—mind EAs are rare—donate on a whim, incentivized by the announcement of the matching, without that they would have donated that money anywhere else with any particularly high probability.
I see another critique to apply with the schemes that have matching “up to a specified level, say $500,000”, and I think you have not mentioned exactly that one explicitly. That additional critique is as follows: If that level of $500k is expected to be reached in due time, then anyone whose donation had been matched before the fund ran dry, has in fact led not to a total donation increase > his personal contribution, but instead in fact to one < than his personal contribution (in the most extreme case 0): because of his donation, the fund has run dry a bit earlier, leaving room for one other person less to donate within the scheme; the total matchmaker contribution remains anyway $500k, but one third person less was incentivized to contribute (because you ‘dried out’ the matching fund earlier). So the matching in reality means your donation has had less impact rather than more, even if you and all donors would not have had other opportunities to donate, i.e. even independently of what I see as one of the main critiques you mention.
Part of your critique is mostly valid in cases where donors have a fixed donation budget and allocate it to the best cause they come across, taking into account a potential leverage factor. I wonder whether instead a lot of the donors—mind EAs are rare—donate on a whim, incentivized by the announcement of the matching, without that they would have donated that money anywhere else with any particularly high probability.
I see another critique to apply with the schemes that have matching “up to a specified level, say $500,000”, and I think you have not mentioned exactly that one explicitly. That additional critique is as follows: If that level of $500k is expected to be reached in due time, then anyone whose donation had been matched before the fund ran dry, has in fact led not to a total donation increase > his personal contribution, but instead in fact to one < than his personal contribution (in the most extreme case 0): because of his donation, the fund has run dry a bit earlier, leaving room for one other person less to donate within the scheme; the total matchmaker contribution remains anyway $500k, but one third person less was incentivized to contribute (because you ‘dried out’ the matching fund earlier). So the matching in reality means your donation has had less impact rather than more, even if you and all donors would not have had other opportunities to donate, i.e. even independently of what I see as one of the main critiques you mention.