The Paxlovid prediction market seems to offer some sort of coordination mechanism for lobbying the FDA. Imagine a professional FDA lobbyist/activist who purchases shares in the FDA approving Paxlovid before the end of the year, and then goes on to launch an energetic lobbying campaign to convince them to do it. This is just a different (and, I think, positive) spin on the “assassination market” concept.
Instead, it’s a sort of “activism market,” a way to indirectly organize payment for public goods even more generally.
Step 1) Convince polymarket to set up a question relevant to the activist work you’d like to accomplish. Make sure there is a concrete benchmark for that activism. How many ships will be backed up at the Port of Long Beach on Jan. 31st, 2022? Will the FDA EUA/approve Paxlovid by Dec. 31, 2021?
Step 2) Buy lots of shares contingent on a positive outcome.
Step 3) Organize an activist campaign to bring about that outcome.
Step 4) Profit!
There are definitely threats and challenges posed by this technology, but I think the positives are exciting as well and worth dwelling on!
We still need to prevent this from becoming an assassination market—we need some mechanism that prevents the equal-and-opposite outcome of a professional FDA lobbyist/activist purchasing shares in the FDA not approving Paxlovid, and then going on to run a campaign to prevent it.
That’s a good point. I think the right frame here is “activation energy” and “catalysis.”
Let’s say that for each uncertain question, the outcome is based on three factors at any given time. One is random chance. Another is potential energy—the most stable possible state. The third is activation energy—the energy input required to move from the current state to a more stable state.
Prediction-market-funded activism seems like it would just provide a catalytic input for change generally, lowering activation energy across the board.
So for the incentive effect to be net good, you’d sort of have to assume that on the whole, the long-run most stable states for social changes are best. You’d also assume that activation energy (incentives, coordination) is the main barrier to getting there from here.
So in the case of Paxlovid, if you assume that the “stable state” is getting it approved—since it’s so effective and necessary—then lowering the activation energy for activism seems likely ultimately to result in speeding its approval. Even if it got anti-Paxlovid activists motivates good, those activists would be pushing uphill.
This optimism is just a gut assumption that underpins my intuitions that prediction markets could be net good in situations like this. But that’s literally all it is at this point—a gut assumption.
I’d be very surprised if people trading in prediction markets then battling it out to steer the FDA wouldn’t be an improvement over current conditions, but I also think there’s a lot of zeroes before that’s something to consider.
It might just be status quo bias or cynicism-driven pattern-matching, but I feel like for any given deadline, paxlovid-is-illegal is more of a “stable state” than paxlovid-is-legal—it feels like it would be easier to lock the general public into ’paxlovid is dangerous/untrustworthy/ineffective” with a campaign against it than it would be to lock the general public into a state of “paxlovid is safe and works and we use it” with a campaign for it, although now I’m actually trying to visualise a world in which paxlovid remains illegal indefinitely in the face of evidence I feel less confident in that cynicism than I did two weeks ago.
The Paxlovid prediction market seems to offer some sort of coordination mechanism for lobbying the FDA. Imagine a professional FDA lobbyist/activist who purchases shares in the FDA approving Paxlovid before the end of the year, and then goes on to launch an energetic lobbying campaign to convince them to do it. This is just a different (and, I think, positive) spin on the “assassination market” concept.
Instead, it’s a sort of “activism market,” a way to indirectly organize payment for public goods even more generally.
Step 1) Convince polymarket to set up a question relevant to the activist work you’d like to accomplish. Make sure there is a concrete benchmark for that activism. How many ships will be backed up at the Port of Long Beach on Jan. 31st, 2022? Will the FDA EUA/approve Paxlovid by Dec. 31, 2021?
Step 2) Buy lots of shares contingent on a positive outcome.
Step 3) Organize an activist campaign to bring about that outcome.
Step 4) Profit!
There are definitely threats and challenges posed by this technology, but I think the positives are exciting as well and worth dwelling on!
We still need to prevent this from becoming an assassination market—we need some mechanism that prevents the equal-and-opposite outcome of a professional FDA lobbyist/activist purchasing shares in the FDA not approving Paxlovid, and then going on to run a campaign to prevent it.
That’s a good point. I think the right frame here is “activation energy” and “catalysis.”
Let’s say that for each uncertain question, the outcome is based on three factors at any given time. One is random chance. Another is potential energy—the most stable possible state. The third is activation energy—the energy input required to move from the current state to a more stable state.
Prediction-market-funded activism seems like it would just provide a catalytic input for change generally, lowering activation energy across the board.
So for the incentive effect to be net good, you’d sort of have to assume that on the whole, the long-run most stable states for social changes are best. You’d also assume that activation energy (incentives, coordination) is the main barrier to getting there from here.
So in the case of Paxlovid, if you assume that the “stable state” is getting it approved—since it’s so effective and necessary—then lowering the activation energy for activism seems likely ultimately to result in speeding its approval. Even if it got anti-Paxlovid activists motivates good, those activists would be pushing uphill.
This optimism is just a gut assumption that underpins my intuitions that prediction markets could be net good in situations like this. But that’s literally all it is at this point—a gut assumption.
I’d be very surprised if people trading in prediction markets then battling it out to steer the FDA wouldn’t be an improvement over current conditions, but I also think there’s a lot of zeroes before that’s something to consider.
It might just be status quo bias or cynicism-driven pattern-matching, but I feel like for any given deadline, paxlovid-is-illegal is more of a “stable state” than paxlovid-is-legal—it feels like it would be easier to lock the general public into ’paxlovid is dangerous/untrustworthy/ineffective” with a campaign against it than it would be to lock the general public into a state of “paxlovid is safe and works and we use it” with a campaign for it, although now I’m actually trying to visualise a world in which paxlovid remains illegal indefinitely in the face of evidence I feel less confident in that cynicism than I did two weeks ago.