Concrete example: in a futarchy system, we have 3 current indices, Ia, Ib, Ic, reflecting desired values, and 3 proposals P1, P2, P3 just passed. In 1 year, you measure the indices and confirm that they increased, which means the proposals succeeded. Problem is, how do you pick which proposals were responsible for the good effects? It could very well be the case that proposal P1 and P2 were outstanding, but P3 was awfully damaging. Since the overall indices increased, you can't detect that P3 was a bad choice, you can only assert the group of proposals did good.


As usually framed, futarchy doesn't involve waiting to see what has the best result when actually implemented. The idea is that you get people to bet in advance about what they think will have the best result, then you implement that.

Imagine GM want to increase their stock price, and they're wondering whether it would help if they fired their CEO. You give people the ability to bet on two different markets: The price in a world where GM did fire their CEO and the price where GM didn't fire their CEO. The simplest way to do this is just to void any bets on the market premised on an outcome that didn't happen. So you can bet on both markets, the one where the CEO was fired and the one where they won't, but your stake for the one that isn't implemented will simply be returned to you.

Set up this way there should be no particular limit to how many propositions you can test simultaneously. However, there may of course be situations where you want to test a bundle of propositions. "Sell the Toyota and buy a Mazda" and "Keep the Toyota and don't buy a Mazda" may be good ideas where "Sell the Toyota and don't buy a Mazda" and "Keep the Toyota and buy a Mazda" are both bad ideas.

In the context of a DAO you would typically have a token with its value taking the place of GM stock. One suggested approach is to enable buying and selling of the token with the proviso, "this sale only happens if proposition P1 is passed". This will set up a market in the token and produce a price, which can then be used to feed back into the decision whether the proposition passes or not. As with GM stock, you can run P2 and P3 in parallel.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.