Happy to describe what we’ve historically done, though we are still figuring out some better policies here, so I expect this to have changed by next round.
As I mentioned in the description of our voting process last time, we have a spreadsheet that has all the applications and other organizations we are considering recommending grants to, in which each fund member can indicate a vote as well as a suggested funding amount (we are still iterating on whether to use votes or a more budget-based system). Whenever there is a potential cause for a conflict of interest, the relevant person (or a separate fund member who suspects a COI) leaves a comment in the relevant row in the spreadsheet with the details of the COI, then the other fund members look at the description of the COI and decide whether it makes sense for that person to withdraw from voting (so far the fund always agreed with the assessment of the individual fund member on whether they should withdraw, but since any fund member has veto power over any of our grant recommendations, I am confident that we would not make grants in which one fund member thinks that a different fund member has a strong COI and we don’t have independent evidence supporting the grant).
We don’t (yet) have a super concrete policy of what counts as a conflict of interest, but I think we’ve historically been quite conservative and flagged all the following things as potential conflicts of interest (this doesn’t mean I am certain that all of the below have been flagged on every occasion, though I think that’s quite likely, just that all of these have historically been flagged) :
Having worked with the relevant organization or people in the past (this usually just gets flagged internally and then added to the writeup, and doesn’t cause someone to withdraw from voting)
Living in the same house/apartment as the person we are granting to (usually makes us hesitant to make a grant just on the basis of the relevant person, so we usually seek additional external feedback to compensate for that)
Being long-time friends with the potential grantee (I expect would just get flagged and added to the writeup)
Being a past or current romantic partner of the potential grantee (I expect this would cause someone to exclude themselves from voting, though I don’t think this ever became relevant. There is one case where a fund member first met and started dating a potential grantee after all the votes had been finalized, but I don’t think there was any undue influence in that case.)
Having some other interpersonal conflict with the relevant person (This usually doesn’t make it into the writeup, but I flagged it on one occasion)
Probably some others, since COIs can arise from all kinds of things
If the votes of the fund member with the potential COI mattered in our final grant decision we’ve incorporated descriptions of those COIs into the final writeups, and sometimes added writeups by people who have less cause for a COI to provide a more neutral source of input (an example of this is the Stag grant, which has a writeup by Alex who has some degree of COI in his relationship to Stag, so it seemed good to add a writeup by me as an additional assessment of the grant).
CEA is currently drafting a more formal policy which is stricter about fund members making recommendations to their own organizations, or organizations closely related to their own organization, but doesn’t cover most of the other things above. We are also currently discussing some more formalized COI policy internally, though I always expect that for the vast majority of potential COI causes we will have to rely on a relatively fuzzy definition, because these can arise from all kinds of different things.
Thank you, this speaks to some tentative concerns I had after reading the grant recommendations. FWIW, I feel that the following was particularly helpful for me deciding how much I trust the decision-making behind grant decisions, how willing I feel to donate to the Fund etc. - I think I would have liked to see this information in the top-level post.
Whenever there is a potential cause for a conflict of interest, the relevant person (or a separate fund member who suspects a COI) leaves a comment in the relevant row in the spreadsheet with the details of the COI, then the other fund members look at the description of the COI and decide whether it makes sense for that person to withdraw from voting.
That’s good to hear! Because a lot of what I described above was a relatively informal procedure, I felt weird putting a lot of emphasis on it in the writeup, but I do agree that it seems like important information for others to have.
I think by next round we will probably have a more formal policy that I would feel more comfortable explicitly emphasizing in the writeup.
Happy to describe what we’ve historically done, though we are still figuring out some better policies here, so I expect this to have changed by next round.
As I mentioned in the description of our voting process last time, we have a spreadsheet that has all the applications and other organizations we are considering recommending grants to, in which each fund member can indicate a vote as well as a suggested funding amount (we are still iterating on whether to use votes or a more budget-based system). Whenever there is a potential cause for a conflict of interest, the relevant person (or a separate fund member who suspects a COI) leaves a comment in the relevant row in the spreadsheet with the details of the COI, then the other fund members look at the description of the COI and decide whether it makes sense for that person to withdraw from voting (so far the fund always agreed with the assessment of the individual fund member on whether they should withdraw, but since any fund member has veto power over any of our grant recommendations, I am confident that we would not make grants in which one fund member thinks that a different fund member has a strong COI and we don’t have independent evidence supporting the grant).
We don’t (yet) have a super concrete policy of what counts as a conflict of interest, but I think we’ve historically been quite conservative and flagged all the following things as potential conflicts of interest (this doesn’t mean I am certain that all of the below have been flagged on every occasion, though I think that’s quite likely, just that all of these have historically been flagged) :
Having worked with the relevant organization or people in the past (this usually just gets flagged internally and then added to the writeup, and doesn’t cause someone to withdraw from voting)
Living in the same house/apartment as the person we are granting to (usually makes us hesitant to make a grant just on the basis of the relevant person, so we usually seek additional external feedback to compensate for that)
Being long-time friends with the potential grantee (I expect would just get flagged and added to the writeup)
Being a past or current romantic partner of the potential grantee (I expect this would cause someone to exclude themselves from voting, though I don’t think this ever became relevant. There is one case where a fund member first met and started dating a potential grantee after all the votes had been finalized, but I don’t think there was any undue influence in that case.)
Having some other interpersonal conflict with the relevant person (This usually doesn’t make it into the writeup, but I flagged it on one occasion)
Probably some others, since COIs can arise from all kinds of things
If the votes of the fund member with the potential COI mattered in our final grant decision we’ve incorporated descriptions of those COIs into the final writeups, and sometimes added writeups by people who have less cause for a COI to provide a more neutral source of input (an example of this is the Stag grant, which has a writeup by Alex who has some degree of COI in his relationship to Stag, so it seemed good to add a writeup by me as an additional assessment of the grant).
CEA is currently drafting a more formal policy which is stricter about fund members making recommendations to their own organizations, or organizations closely related to their own organization, but doesn’t cover most of the other things above. We are also currently discussing some more formalized COI policy internally, though I always expect that for the vast majority of potential COI causes we will have to rely on a relatively fuzzy definition, because these can arise from all kinds of different things.
Thank you for a very thorough and transparent reply!
Thank you, this speaks to some tentative concerns I had after reading the grant recommendations. FWIW, I feel that the following was particularly helpful for me deciding how much I trust the decision-making behind grant decisions, how willing I feel to donate to the Fund etc. - I think I would have liked to see this information in the top-level post.
That’s good to hear! Because a lot of what I described above was a relatively informal procedure, I felt weird putting a lot of emphasis on it in the writeup, but I do agree that it seems like important information for others to have.
I think by next round we will probably have a more formal policy that I would feel more comfortable explicitly emphasizing in the writeup.