I’ll ask whoever runs the utility.farm site to update my piece cited in this with a note that the cost-effectiveness estimates might be based on bad estimates of cat impact.
Additionally, my cost-effectiveness estimates were only for the US—it is probably most cost-effective to work on cat predation in countries like the UK where a much higher percentage of outdoor cats are owned.
I find the comments about rodents/birds interesting, but mostly irrelevant to the discussion of cat predation, and find framing addressing cat predation and improving rodent welfare as competing aims very strange. I’m going to refer to rodents below, but it could apply to any animals killed by cats. It doesn’t seem obvious that the causal chain we should care about is stopping cat predation causing painful rodent deaths; instead, we should consider both rodenticides causing painful rodent deaths and cat predation causing painful rodent deaths to be important issues.
For there to be a coherent argument to not address cat predation, you would need to demonstrate not only that rodenticides more painful than death via cats, but have a picture of the average rodent’s life after the moment they might have been killed by a cat. Since any rodent who is killed by a cat definitionally would have lived longer had it been killed by a rodenticide instead, the rodent is going to accumulate further positive and negative experiences during its life before being killed. Even if rodenticides are twice as painful, it seems reasonable to expect a prolonged life to often be good, and outweigh that.
Regardless, this doesn’t seem like an argument against addressing cat predation—its an argument that the most effective way to address rodent suffering might be both addressing cat predation and addressing painful rodenticides.
A human analogy might be: we shouldn’t address malaria because someone dying from malaria, if saved, might die from another more painful disease later. I think what follows from that is that we should address malaria and the other painful disease. Not that we should let malaria kill the person / say that it is unclear if malaria is good or bad. Addressing malaria is clearly good, as is reducing cat predation. There might be unintended effects of both that also need to be addressed. But it doesn’t mean that addressing those things has an unclear sign.
My point is, cat predation is clearly bad for rodents. Rodenticides are also clearly bad. We should probably address both things, and be aware of the effect of only addressing one, but not that we shouldn’t address either. Broadly, this seems to apply to wild animal welfare issues in general—the downstream effects are really really complicated, but by doing monitoring during interventions to see unanticipated effects, and addressing those as they come up, we probably make more progress than just pointing out the complexity.
I guess the question that is raised is, how far down should we care about downstream effects from interventions, as opposed to just monitoring interventions and addressing effects as they arise.
The position of this piece is that, given we have no knowledge regarding the central tendency and distribution of rodent well-being, all we can do realistically is compare the quality of deaths. Whether you view an earlier death as a positive or a negative depends on both whether you assume rodent lives are on average net positive or net negative, and whether you assume the subsequent numerous offspring are likely to experience net positive or net negative lives. We are agnostic on this, but do point out that the length of time suffering due to poisoning or starvation will be longer than the suffering due to cat predation. However, one point we raised regarding outdoor cat presence is of course that fear-based vicinity control could reduce total deaths due to both poisoning and predation (although we certainly accept that this is this is not a simple approach Bedoya-Perez et al 2019, Krijger et al. 2017).
These points aside, the reason that we frame cat predation and rodenticide use as alternatives is that if you prioritize human interests and welfare above wildlife interests (and again we are agnostic here) then you will absolutely need to come up with an alternative method of control. Commensal rodent populations have large economic and human health impacts (e.g. Meerburg et al. 2009, Stenseth et al. 2003). (As an aside, given the tremendous reproductive output, you will also need controls if you consider rodent starvation something to be averted). With respect to downstream effects, cat predation and fear generation are methods of control that may be under appreciated and which at least don’t have the far-reaching chemical contamination issues that rodenticides do.
Lit cited
Bedoya-Perez, M. A., Smith, K. L., Kevin, R. C., Luo, J. L., Crowther, M. S., & McGregor, I. S. (2019). Parameters that affect fear responses in rodents and how to use them for management. Frontiers in Ecology and Evolution, 7, 136. https://doi.org/10.3389/fevo.2019.00136
Krijger, I. M., Belmain, S. R., Singleton, G. R., Groot Koerkamp, P. W., & Meerburg, B. G. (2017). The need to implement the landscape of fear within rodent pest management strategies. Pest management science, 73(12), 2397-2402. https://doi.org/10.1002/ps.4626
Meerburg, B. G., Singleton, G. R., & Kijlstra, A. (2009). Rodent-borne diseases and their risks for public health. Critical reviews in microbiology, 35(3), 221-270. https://doi.org/10.1080/10408410902989837
Stenseth, N. C., Leirs, H., Skonhoft, A., Davis, S. A., Pech, R. P., Andreassen, H. P., … & Zhang, Z. (2003). Mice, rats, and people: the bio‐economics of agricultural rodent pests. Frontiers in Ecology and the Environment, 1(7), 367-375. https://doi.org/10.1890/1540-9295(2003)001[0367:MRAPTB]2.0.CO;2
Thanks for the detailed response. I think I disagree in a sort of principled way with particular kinds of approaches to downstream effects, in part because I think it could just turn into an endless game of trying to figure out how things could turn out poorly, as opposed to a model where we address both rodenticides and cat predation (though I recognize I am stubbornly resisting you all trying to do prioritization, which might not be a good idea given the name of your organization).
Regardless, I’m drafting a new intro section for my cost-effectiveness updates linking to your updated numbers. Thanks again for doing the analysis!
I’ll ask whoever runs the utility.farm site to update my piece cited in this with a note that the cost-effectiveness estimates might be based on bad estimates of cat impact.
Additionally, my cost-effectiveness estimates were only for the US—it is probably most cost-effective to work on cat predation in countries like the UK where a much higher percentage of outdoor cats are owned.
I find the comments about rodents/birds interesting, but mostly irrelevant to the discussion of cat predation, and find framing addressing cat predation and improving rodent welfare as competing aims very strange. I’m going to refer to rodents below, but it could apply to any animals killed by cats. It doesn’t seem obvious that the causal chain we should care about is stopping cat predation causing painful rodent deaths; instead, we should consider both rodenticides causing painful rodent deaths and cat predation causing painful rodent deaths to be important issues.
For there to be a coherent argument to not address cat predation, you would need to demonstrate not only that rodenticides more painful than death via cats, but have a picture of the average rodent’s life after the moment they might have been killed by a cat. Since any rodent who is killed by a cat definitionally would have lived longer had it been killed by a rodenticide instead, the rodent is going to accumulate further positive and negative experiences during its life before being killed. Even if rodenticides are twice as painful, it seems reasonable to expect a prolonged life to often be good, and outweigh that.
Regardless, this doesn’t seem like an argument against addressing cat predation—its an argument that the most effective way to address rodent suffering might be both addressing cat predation and addressing painful rodenticides.
A human analogy might be: we shouldn’t address malaria because someone dying from malaria, if saved, might die from another more painful disease later. I think what follows from that is that we should address malaria and the other painful disease. Not that we should let malaria kill the person / say that it is unclear if malaria is good or bad. Addressing malaria is clearly good, as is reducing cat predation. There might be unintended effects of both that also need to be addressed. But it doesn’t mean that addressing those things has an unclear sign.
My point is, cat predation is clearly bad for rodents. Rodenticides are also clearly bad. We should probably address both things, and be aware of the effect of only addressing one, but not that we shouldn’t address either. Broadly, this seems to apply to wild animal welfare issues in general—the downstream effects are really really complicated, but by doing monitoring during interventions to see unanticipated effects, and addressing those as they come up, we probably make more progress than just pointing out the complexity.
I guess the question that is raised is, how far down should we care about downstream effects from interventions, as opposed to just monitoring interventions and addressing effects as they arise.
The position of this piece is that, given we have no knowledge regarding the central tendency and distribution of rodent well-being, all we can do realistically is compare the quality of deaths. Whether you view an earlier death as a positive or a negative depends on both whether you assume rodent lives are on average net positive or net negative, and whether you assume the subsequent numerous offspring are likely to experience net positive or net negative lives. We are agnostic on this, but do point out that the length of time suffering due to poisoning or starvation will be longer than the suffering due to cat predation. However, one point we raised regarding outdoor cat presence is of course that fear-based vicinity control could reduce total deaths due to both poisoning and predation (although we certainly accept that this is this is not a simple approach Bedoya-Perez et al 2019, Krijger et al. 2017).
These points aside, the reason that we frame cat predation and rodenticide use as alternatives is that if you prioritize human interests and welfare above wildlife interests (and again we are agnostic here) then you will absolutely need to come up with an alternative method of control. Commensal rodent populations have large economic and human health impacts (e.g. Meerburg et al. 2009, Stenseth et al. 2003). (As an aside, given the tremendous reproductive output, you will also need controls if you consider rodent starvation something to be averted). With respect to downstream effects, cat predation and fear generation are methods of control that may be under appreciated and which at least don’t have the far-reaching chemical contamination issues that rodenticides do.
Lit cited
Bedoya-Perez, M. A., Smith, K. L., Kevin, R. C., Luo, J. L., Crowther, M. S., & McGregor, I. S. (2019). Parameters that affect fear responses in rodents and how to use them for management. Frontiers in Ecology and Evolution, 7, 136. https://doi.org/10.3389/fevo.2019.00136
Krijger, I. M., Belmain, S. R., Singleton, G. R., Groot Koerkamp, P. W., & Meerburg, B. G. (2017). The need to implement the landscape of fear within rodent pest management strategies. Pest management science, 73(12), 2397-2402. https://doi.org/10.1002/ps.4626
Meerburg, B. G., Singleton, G. R., & Kijlstra, A. (2009). Rodent-borne diseases and their risks for public health. Critical reviews in microbiology, 35(3), 221-270. https://doi.org/10.1080/10408410902989837
Stenseth, N. C., Leirs, H., Skonhoft, A., Davis, S. A., Pech, R. P., Andreassen, H. P., … & Zhang, Z. (2003). Mice, rats, and people: the bio‐economics of agricultural rodent pests. Frontiers in Ecology and the Environment, 1(7), 367-375. https://doi.org/10.1890/1540-9295(2003)001[0367:MRAPTB]2.0.CO;2
Thanks for the detailed response. I think I disagree in a sort of principled way with particular kinds of approaches to downstream effects, in part because I think it could just turn into an endless game of trying to figure out how things could turn out poorly, as opposed to a model where we address both rodenticides and cat predation (though I recognize I am stubbornly resisting you all trying to do prioritization, which might not be a good idea given the name of your organization).
Regardless, I’m drafting a new intro section for my cost-effectiveness updates linking to your updated numbers. Thanks again for doing the analysis!