4 ms·

The funny thing is that the Environmental metric is the only objective one while the Social and Governance ones can be bullshitted and manipulated. Those three

The funny thing is that the Environmental metric is the only objective one while the Social and Governance ones can be bullshitted and manipulated. Those three things should have never been coupled. It’s all a scam.

gnramires 3y ago
I dunno, I think it could be combined in smarter ways, e.g. min(E,S,G) (or some kind of soft-min)[1]. Or give different weights for more uncertain measures. I think there's a very large amount of knowledge (in various fields) around evaluating those things, and we could have much more if we put some effort; also, we don't necessarily need to make everything 100% based on a small set of metrics. You could have some kind of adaptive evaluation that tries to find the most impactful events for that organization (e.g. trying to track down pollutants and quantify their impact, trying to track down social impacts and quantify them, etc.). In medical and charity fields for example there is QALYs (which do have serious problems still!), but are at least a reference point for impact -- to given an example; in comparing development we have HDI; etc.. Overall, I think trying to improve those things in a systematic way is the only real way to improve them. I want to know how well each company is doing in various areas. We are currently mostly blind to that. And we should be mindful not to let the ones trying to game and undermine the systems win, and sincerely look at impact in all areas. [1] Ideally, I think a metric would be able to measure something like a 'meaning of life difference' from different choices (and assign a corresponding metric for say an organization based on certain counterfactuals). Of course that's too difficult in general, but we can strive for example to keep the planet healthy and establish some standard "unit" for a healthy society (could be translated to money, but there could be some issues with that), establishing a trade-off say between say saving someone's life now (through a health intervention), and saving lives in the future by improving planetary conditions (with less pollution). Those things are perhaps surprisingly comparable (and surprisingly linear/additive as well). There are mentions of Goodhart's Law and it certainly applies somewhat (if we tried simple, naive metrics): but Goodhart's Law doesn't apply when there are real, smart people doing the evaluation, in a dynamic way, using quantitative tools sensibly. And finally there are choices perhaps no such evaluation could capture, questions about what future do we want for ourselves in a broader cultural and artistic sense (which is why in the end freedom to support what you want as an individual is important). I really think this is going to play a significant part in how we address many large scale issues!

gnramires 3y ago

I dunno, I think it could be combined in smarter ways, e.g. min(E,S,G) (or some kind of soft-min)[1]. Or give different weights for more uncertain measures. I think there's a very large amount of knowledge (in various fields) around evaluating those things, and we could have much more if we put some effort; also, we don't necessarily need to make everything 100% based on a small set of metrics. You could have some kind of adaptive evaluation that tries to find the most impactful events for that organization (e.g. trying to track down pollutants and quantify their impact, trying to track down social impacts and quantify them, etc.). In medical and charity fields for example there is QALYs (which do have serious problems still!), but are at least a reference point for impact -- to given an example; in comparing development we have HDI; etc.. Overall, I think trying to improve those things in a systematic way is the only real way to improve them. I want to know how well each company is doing in various areas. We are currently mostly blind to that. And we should be mindful not to let the ones trying to game and undermine the systems win, and sincerely look at impact in all areas. [1] Ideally, I think a metric would be able to measure something like a 'meaning of life difference' from different choices (and assign a corresponding metric for say an organization based on certain counterfactuals). Of course that's too difficult in general, but we can strive for example to keep the planet healthy and establish some standard "unit" for a healthy society (could be translated to money, but there could be some issues with that), establishing a trade-off say between say saving someone's life now (through a health intervention), and saving lives in the future by improving planetary conditions (with less pollution). Those things are perhaps surprisingly comparable (and surprisingly linear/additive as well). There are mentions of Goodhart's Law and it certainly applies somewhat (if we tried simple, naive metrics): but Goodhart's Law doesn't apply when there are real, smart people doing the evaluation, in a dynamic way, using quantitative tools sensibly. And finally there are choices perhaps no such evaluation could capture, questions about what future do we want for ourselves in a broader cultural and artistic sense (which is why in the end freedom to support what you want as an individual is important). I really think this is going to play a significant part in how we address many large scale issues!