post by kbog
Comments sorted by top scores.
comment by kbog ·
2020-06-11T02:58:51.687Z · EA(p) · GW(p)
I just noticed this post [EA · GW] and the ensuing discussion. I want to share a model I recently made which seeks to answer the question: are these protests beneficial or harmful.
- The expected deaths caused by COVID spread outnumber the expected lives saved from reducing police brutality by a factor of 16.
- If we adjust for QALYs (COVID mainly kills older folks), the COVID mortality is still worse than the reduction in police killings, though only by a factor of 5.
- When I estimate a general positive impact of these protests upon America's political system - specifically, that they'll increase Democratic voteshare this November - it seems that the protests are neutral as far as American citizens are concerned, but (more importantly of course) positive when we include foreigners and animals.
I want to say upfront that this doesn't mean I endorse the protests, I still feel a bit negative about them due to the prisoner's dilemma at play (as in Larks' highly upvoted comment in the other thread, I also came up with the same point).Replies from: MichaelStJules
↑ comment by MichaelStJules ·
2020-06-12T02:08:55.775Z · EA(p) · GW(p)
I haven't looked into this in any detail myself, but I wonder if other benefits for American POC could be more important than the prevented deaths, e.g. poverty, mental health and other crisis services, plea bargaining, imprisonment and sentencing. I think this could open a broader conversation about criminal justice. Some of this could come from a Dem presidency, but it might be addressed at state or more local levels, too.
comment by kbog ·
2020-06-11T07:45:47.601Z · EA(p) · GW(p)
Shouldn't we collect a sort of encyclopedia or manual of organizational best practices, to help EA organizations? A combination of research, and things we have learned? Replies from: Prabhat Soni
↑ comment by Prabhat Soni ·
2020-06-28T18:44:08.654Z · EA(p) · GW(p)
Why couldn't a manual of organizational best practices from non-EA organisations (I'm guessing there are probably many such manuals or other ways of communicating best practices) suffice? Which areas would it be unable to cover when applied directly to EA organisations? Are these areas particularly important to cover?