Do you estimate cost-effectiveness based on guesses for the intensity of the 4 categories of pain defined by the Welfare Footprint Project (WFP), as Ido, and Ambitious Impact (AIM) does? If so, which values are you using?
I think AIM’s values greatly underestimate the intensity of severe pain. Feel free to ask Vicky Cox for the doc with my suggestions for improvement.
I’d refer you to an answer we gave in a previous post about how the fund has historically relied on a range of factors to judge marginal cost-effectiveness for the majority of grants and it’s less often the case we have the evidence and the grant size merits a formal cost-effectiveness model. Having said that, we are currently trialing different approaches to cost-effectiveness modeling as that becomes a more standard feature of our deep evaluations (see more in our FAQs on our grantmaking process) and making explicit BOTECs a required part of evaluations. Among these, models we’re investigating include how a range of different pain category intensities weightings (as described in Grilo 2024, Ryba 2024, Schuck et al. 2024) could affect our cost-effectiveness estimates. There are reasonable grounds to put some credence in the most severe harms causing farmed animals at least as much disutility as the longest-lasting harms they experience (McAuliffe and Shriver 2023, also see Parra 2024, Ryba 2023).We make sure to note in the evaluation if the overall assessment would hinge on such a consideration (or other more philosophical/​ fundamental questions where people have reasonable disagreements) to guard against being systematically biased towards one perspective. However, in practice, this may only be a crux for a handful of grant applications (e.g., those focused on pre-slaughter stunning) Often to make a grant decision we don’t need to get a precise estimate down to the exact total hours of intensity-adjusted pain, just what would one need to believe for this grant to be at least competitive with other opportunities and does that seem like a reasonable belief to hold.
I don’t think they’re doing explicit cost-effectiveness analyses at this level of detail for the grants they’re assessing. They don’t have the time (consider how many hours they have per grant vetted), and the evidence to make those analyses anything more than multiplying a bunch of guesses together just won’t exist for many applications. Instead, they’ll be looking at other forms of evidence, like the strength of the theory of change, the strength of the team, the quality of thinking displayed in the team’s plan and their track record
Thanks, Aidan. You may well be right about what they are doing, but I think a basic version would take very little time, and still be helpful. They would only need to make one estimation of the DALYs averted per animal-year improved or animal helped for each type of welfare improvement, not one estimation per grant. They could make copies of my sheets for broiler welfare and cage-free campaigns, and for improving shrimp slaughter, although I am sure they could do similar sheets in no time. Then they could have these welfare improvements in mind while making guesses for others for which there is no data on the time in pain.
Do you estimate cost-effectiveness based on guesses for the intensity of the 4 categories of pain defined by the Welfare Footprint Project (WFP), as I do, and Ambitious Impact (AIM) does? If so, which values are you using?
I think AIM’s values greatly underestimate the intensity of severe pain. Feel free to ask Vicky Cox for the doc with my suggestions for improvement.
Thanks for the question!
I’d refer you to an answer we gave in a previous post about how the fund has historically relied on a range of factors to judge marginal cost-effectiveness for the majority of grants and it’s less often the case we have the evidence and the grant size merits a formal cost-effectiveness model. Having said that, we are currently trialing different approaches to cost-effectiveness modeling as that becomes a more standard feature of our deep evaluations (see more in our FAQs on our grantmaking process) and making explicit BOTECs a required part of evaluations. Among these, models we’re investigating include how a range of different pain category intensities weightings (as described in Grilo 2024, Ryba 2024, Schuck et al. 2024) could affect our cost-effectiveness estimates.
There are reasonable grounds to put some credence in the most severe harms causing farmed animals at least as much disutility as the longest-lasting harms they experience (McAuliffe and Shriver 2023, also see Parra 2024, Ryba 2023).We make sure to note in the evaluation if the overall assessment would hinge on such a consideration (or other more philosophical/​ fundamental questions where people have reasonable disagreements) to guard against being systematically biased towards one perspective. However, in practice, this may only be a crux for a handful of grant applications (e.g., those focused on pre-slaughter stunning)
Often to make a grant decision we don’t need to get a precise estimate down to the exact total hours of intensity-adjusted pain, just what would one need to believe for this grant to be at least competitive with other opportunities and does that seem like a reasonable belief to hold.
I don’t think they’re doing explicit cost-effectiveness analyses at this level of detail for the grants they’re assessing. They don’t have the time (consider how many hours they have per grant vetted), and the evidence to make those analyses anything more than multiplying a bunch of guesses together just won’t exist for many applications. Instead, they’ll be looking at other forms of evidence, like the strength of the theory of change, the strength of the team, the quality of thinking displayed in the team’s plan and their track record
Thanks, Aidan. You may well be right about what they are doing, but I think a basic version would take very little time, and still be helpful. They would only need to make one estimation of the DALYs averted per animal-year improved or animal helped for each type of welfare improvement, not one estimation per grant. They could make copies of my sheets for broiler welfare and cage-free campaigns, and for improving shrimp slaughter, although I am sure they could do similar sheets in no time. Then they could have these welfare improvements in mind while making guesses for others for which there is no data on the time in pain.