That assumes that “further research” will reduce these confidence intervals significantly, which I am skeptical of.

You could fund 1000 PostDocs for 1000 years each to study “why is there something rather than nothing” or “is one person’s perception of blue the same as another’s” and it’s no given that you’ll get closer to an answer.

You can’t bake-in something as unpredictable as how movements and counter-movements evolve and interact.

We need to be more open to uncertainty and consider unexpected ways in which our best laid plans may go astray. Animal Welfare is rife with these uncertainties.

This seems very ungenerous to the global health space:

• Malaria nets are based on RCTs. Here's a Cochrane review of 22 RCTs:
• Against Malaria Foundation does quite intensive monitoring of uptake (not perfect, but you're implying none)
• New Incentives is based on an RCT and also monitors many metrics
• Malaria consortium is also based on RCTs and does monitoring
• Seva and Fred Hollows track and publish their cataract surgery numbers
• Innovations for Poverty Action's main purpose is to trial interventions and measure them

studies of the effectiveness of the types of interventions these charities use are generalized, with adjustments for context

That is how RCTs work. You can't have a separate RCT for every situation unfortunately.

I wouldn't advocate giving $100M to Make A Wish just for optics.

But you shouldn't ignore optics, because it affects tractability and can have downstream effects on other parts of the movement.

In a decision between two options where it's ambiguous which is better (global health vs animal welfare) but one has better optics, it is particularly relevant.

I think we agree: the massive uncertainty in the utility calculus approach to this problem could go either way and so it tells us nothing.

In the end we're forced to fall back on our moral intuitions like: "harpooning whale feels bad" and comparative arguments like: "well if you wouldn't suffocate your dog, how can you pay someone to suffocate a pig?". This is the only feasible approach.

Public support is important for getting things done actually. It's affects tractability.

In the case where it's ambiguous which of options A and B are better, but they have different levels of public support, it becomes an important consideration.

I didn't say universal or 50% support. Many women were against, many men were for. My point is that it had a stronger support base than shrimp welfare before we tried to regulate it.

The idea that you can go regulating without considering public support/resistance is silly

Velocity vs displacement

Keeping the public on side is actually quite important for getting things done.

Backlash against the thing you’re trying to promote blows out costs, making the plan less cost-effective

50% of people are women so I think women’s suffrage had a pretty strong support base before it was made law. Similar story for your other examples I think: build support, then laws. Abolition seems like an example of where a counter-movement blew out the cost of change a lot.

The definition of Fermi estimate linked in this post defines a Fermi estimate as aiming to be within 1 magnitude of true. Given just the Rethink Priorities welfare range estimates span several magnitudes (infinite really, given lower bound is 0), this at least is incorrect.

This sort of chaining of EV calculations is common on this forum. I think it's counterproductive. Show the confidence intervals and it becomes clear that the result is as good as "I have no idea", which is a fine thing to say. Just say that.