Add new comment

On minimizing the number of people setting strategy and making decisions:

A potentially useful frame here is one used in the IC and in some business analysis: separating decision making from decision support. You want to analyze any given problem through multiple lenses, and often this work is detailed enough that you want to divy up the work via said lenses. A concrete example is interviewing experts, first principles reasoning, structured probability estimates (scenario analysis) as you mention in discussion of plausible upsides and downsides), reference class forecasting, and Red Teaming might all be things that are assigned to separate people, with their outputs then flowing to the decision maker who ideally will follow best practices in judgment aggregation.

On strategy, priorities and grants for a given cause are largely determined by the single person who is most informed about the cause:

The findings on judgmental bootstrapping in the forecasting literature implies we should be skeptical when an expert is unable to convert their analysis of a domain into an additive linear model that can then easily be shared with others. I’m curious if such models are generated and shared when projects require coordination between multiple researchers, and if not, what the major bottleneck to piloting that would be. Having the affordance of a google sheets template for them for instance could increase decision quality if it causes more and better models to get created and shared on the margin.

On support strong leadership with no strings (or minimal strings) attached, rather than supporting unremarkable people/organizations to carry out plans that appeal to us:

Investing in teams vs ideas seems correct given the reference class of private investing. The potential leak in porting from that reference class is the motivation structure of the teams being dissimilar in the non-profit world vs the private world. It seems difficult to predict in advance which teams will be able to maintain high levels of effectiveness without the same remuneration levers. The traits/tendencies/world views underlying motivation in the non-profit world seem more complex and potentially fragile, though I don’t have any base rate information on that. Might be worth investigating (stability of executive management at non-profits over time?)