Not sure that’s entirely true (I think it’s very interesting). I feel like the grantmaking process is not top→down, but bottom→up→top→down (someone has an idea, they start working on it in their free time (this part is optional), they apply, that thing gets evaluated, they get rejected/receive some money). I think in historical command economies the first/second part of was kind of rare.
Ok this is a solid point and made me slightly reconsider.
However, I question both how much grant making actually is bottom up and also even if it is how different this is from historical command economies.
Depending on the grant makers, I feel like there is a range between “has already decided more or less the projects the want to fund” and “funding anything that seems promising under x goals”, where the former isn’t really “taking suggestions”.
More importantly, while I agree the bottom up thing was rare, I don’t think the issue with command economies is that they don’t gather any information from the crowd, it’s that the information they have available is funneled through human bias and is nontransparent. Would a system of town halls in which command economy leaders had a chance to listen to the complaints of the citizens fix the command economy? This just seems like a worse version of markets.
By the same logic I think impact markets, while they may have a long learning curve, will clearly be superior to what we are currently doing. The main reason I see us not doing this is because you have to actually specify the currency, which would be specifying a moral world view, which would shatter the communities nebulous ethical agnosticism.
Re Coase and Gwern’s post, I think it gives my whole point even more firepower. I feel like the point is you only need one logical feedback mechanism to create evolution. The rest of the system can be nearly random or very nonlogical. But there is no feedback mechanism. We will never be able to tease out the impact of 99% of the things we are currently doing. If grant makers are totalitarians like the firms, where is the analogous feedback loop to money? (without RCTs up the wazoo).
Grant-making as we currently do it seems pretty analogous to a command economy.
Not sure that’s entirely true (I think it’s very interesting). I feel like the grantmaking process is not top→down, but bottom→up→top→down (someone has an idea, they start working on it in their free time (this part is optional), they apply, that thing gets evaluated, they get rejected/receive some money). I think in historical command economies the first/second part of was kind of rare.
See also Coase’s Theory of the Firm and Evolution as a Backstop to Reinforcement Learning.
Ok this is a solid point and made me slightly reconsider.
However, I question both how much grant making actually is bottom up and also even if it is how different this is from historical command economies.
Depending on the grant makers, I feel like there is a range between “has already decided more or less the projects the want to fund” and “funding anything that seems promising under x goals”, where the former isn’t really “taking suggestions”.
More importantly, while I agree the bottom up thing was rare, I don’t think the issue with command economies is that they don’t gather any information from the crowd, it’s that the information they have available is funneled through human bias and is nontransparent. Would a system of town halls in which command economy leaders had a chance to listen to the complaints of the citizens fix the command economy? This just seems like a worse version of markets.
By the same logic I think impact markets, while they may have a long learning curve, will clearly be superior to what we are currently doing. The main reason I see us not doing this is because you have to actually specify the currency, which would be specifying a moral world view, which would shatter the communities nebulous ethical agnosticism.
Re Coase and Gwern’s post, I think it gives my whole point even more firepower. I feel like the point is you only need one logical feedback mechanism to create evolution. The rest of the system can be nearly random or very nonlogical. But there is no feedback mechanism. We will never be able to tease out the impact of 99% of the things we are currently doing. If grant makers are totalitarians like the firms, where is the analogous feedback loop to money? (without RCTs up the wazoo).