Initially I thought it was inappropriate to discuss GiveWell’s executive compensation, but given that: (a) it’s a nonprofit and therefore accountable to the public; and (b) GiveWell publishes board meeting audio, suggesting that it wants public feedback; I believe my concern here is worth raising.
I’ve listened to the audio from both this board meeting and the previous board meeting where they discuss raising the executive salaries at GiveWell; I still don’t really understand why the board found it reasonable to raise the salaries. I only hear two arguments presented:
Other CEO’s make this much.
If Holden or Elie left, they would be expensive to replace.
(2) seems true although fairly minor, and Holden says he doesn’t think this is a major issue. (1) strikes me as a non-sequitur. GiveWell is not trying to hire people to fill the CEO roles and probably will not be trying to do so for the medium-term future so it doesn’t matter whether the salaries are competitive.
Fundamentally, an organization should use salaries to incentivize people to work there and enable the organization to achieve its goals. I absolutely buy that GiveWell/Open Phil should pay high salaries to industry experts who might otherwise work at well-paying for-profit firms. But I don’t believe this makes sense for the executives. If they will quit working at GiveWell if they’re making only $150K, or they will become sufficiently demoralized, then the board could probably justify raising their salaries to $175K. But this seems unlikely and it’s not really discussed at the board meetings, so I don’t understand why it’s reasonable to raise their salaries.
Important disclaimer: I’m not saying that Holden and Elie are not producing $175K of value; they easily produce over $1 million of value per year. Instead, I’m saying it doesn’t make sense to pay people based on what they “deserve”, but rather based on what’s needed to incentivize them to do good work.
Analogously, many startup founders take low salaries while paying their early employees market rates, because they don’t need high salaries to incentivize themselves to do good work. GiveWell should go even further in this direction because its purpose is improving the world, not making a profit.
That’s a fairly decent analogy but I don’t think it quite works. Investing markets are much more efficient than charity markets; if GiveWell didn’t exist, most GiveWell donors probably would have given their money to charities less than half as good. So of the money moved, (conservatively) 50% of that is additional value added, and subtracting 4.6% leaves about 45% value added. Whereas it’s quite rare for investment managers to earn a 5 percentage point premium over the market.
I do tend to think other actors could have produced charity recommendations that are about as good as GiveWell’s. Indeed, I believe Giving What We Can’s recommendations have been on par, and Animal Charity Evaluators’ have been much better (although ACE gets an advantage by focusing on a more cost-effective cause). However, GiveWell has done exceptionally well at attracting funding—no one else has come anywhere close to this, and I tend to think that the ability to attract this much funding is very rare.
ACE’s recommendations may well be much better than Givewell if the general case for animal welfare >> global poverty in terms of cause area. Yet this may be less helpful (a hypothetical far future metacharity I set up now and recommend FHI or whatever might be far better than either by a similar sort of argument).
I think for overall quality of work, Givewell has done much better than ACE, if only by virtue of having far more resources to throw at the issue. In terms of rigour, Givewell probably do much better than anyone.
Perhaps an appropriate metric would be something like quality per full-time equivalent or similar. Then things are pretty murky: GWWC’s track record of recommendations seems about on par with Givewell (I’m doing some work on this at the moment), which is surprising given GWWC has had far fewer research staff and money over this period. This may imply GWWC has been more effective, or that diminishing returns kick in quite early, or something else. The same story may apply to ACE, but ‘track records’ in diverse cause areas are hard.
Initially I thought it was inappropriate to discuss GiveWell’s executive compensation, but given that: (a) it’s a nonprofit and therefore accountable to the public; and (b) GiveWell publishes board meeting audio, suggesting that it wants public feedback; I believe my concern here is worth raising.
I’ve listened to the audio from both this board meeting and the previous board meeting where they discuss raising the executive salaries at GiveWell; I still don’t really understand why the board found it reasonable to raise the salaries. I only hear two arguments presented:
Other CEO’s make this much.
If Holden or Elie left, they would be expensive to replace.
(2) seems true although fairly minor, and Holden says he doesn’t think this is a major issue. (1) strikes me as a non-sequitur. GiveWell is not trying to hire people to fill the CEO roles and probably will not be trying to do so for the medium-term future so it doesn’t matter whether the salaries are competitive.
Fundamentally, an organization should use salaries to incentivize people to work there and enable the organization to achieve its goals. I absolutely buy that GiveWell/Open Phil should pay high salaries to industry experts who might otherwise work at well-paying for-profit firms. But I don’t believe this makes sense for the executives. If they will quit working at GiveWell if they’re making only $150K, or they will become sufficiently demoralized, then the board could probably justify raising their salaries to $175K. But this seems unlikely and it’s not really discussed at the board meetings, so I don’t understand why it’s reasonable to raise their salaries.
Important disclaimer: I’m not saying that Holden and Elie are not producing $175K of value; they easily produce over $1 million of value per year. Instead, I’m saying it doesn’t make sense to pay people based on what they “deserve”, but rather based on what’s needed to incentivize them to do good work.
Analogously, many startup founders take low salaries while paying their early employees market rates, because they don’t need high salaries to incentivize themselves to do good work. GiveWell should go even further in this direction because its purpose is improving the world, not making a profit.
-
That’s a fairly decent analogy but I don’t think it quite works. Investing markets are much more efficient than charity markets; if GiveWell didn’t exist, most GiveWell donors probably would have given their money to charities less than half as good. So of the money moved, (conservatively) 50% of that is additional value added, and subtracting 4.6% leaves about 45% value added. Whereas it’s quite rare for investment managers to earn a 5 percentage point premium over the market.
-
I do tend to think other actors could have produced charity recommendations that are about as good as GiveWell’s. Indeed, I believe Giving What We Can’s recommendations have been on par, and Animal Charity Evaluators’ have been much better (although ACE gets an advantage by focusing on a more cost-effective cause). However, GiveWell has done exceptionally well at attracting funding—no one else has come anywhere close to this, and I tend to think that the ability to attract this much funding is very rare.
‘Betterness’ is challenging here.
ACE’s recommendations may well be much better than Givewell if the general case for animal welfare >> global poverty in terms of cause area. Yet this may be less helpful (a hypothetical far future metacharity I set up now and recommend FHI or whatever might be far better than either by a similar sort of argument).
I think for overall quality of work, Givewell has done much better than ACE, if only by virtue of having far more resources to throw at the issue. In terms of rigour, Givewell probably do much better than anyone.
Perhaps an appropriate metric would be something like quality per full-time equivalent or similar. Then things are pretty murky: GWWC’s track record of recommendations seems about on par with Givewell (I’m doing some work on this at the moment), which is surprising given GWWC has had far fewer research staff and money over this period. This may imply GWWC has been more effective, or that diminishing returns kick in quite early, or something else. The same story may apply to ACE, but ‘track records’ in diverse cause areas are hard.