My point is that people would have an incentive to bet for the fire and then cause the fire, so once a bet is published chances are that the fire will happen.
There would need to be sufficient market depth to pay someone to act. So we couldn’t just publish $0.01 bets on everything and watch the town burn. “Coasian” solutions like this are limited by transaction costs.
In practice, the owner of the datacenter would need to either eliminate the externality or watch it burn. If the datacenter is worth more than its external cost, then they can reimburse their victims or pay for cleanup. If the revenue from the datacenter is less than its external costs, then it’s a net loss for society and ought to be shut down.
My point is that people would have an incentive to bet for the fire and then cause the fire, so once a bet is published chances are that the fire will happen.
There would need to be sufficient market depth to pay someone to act. So we couldn’t just publish $0.01 bets on everything and watch the town burn. “Coasian” solutions like this are limited by transaction costs.
In practice, the owner of the datacenter would need to either eliminate the externality or watch it burn. If the datacenter is worth more than its external cost, then they can reimburse their victims or pay for cleanup. If the revenue from the datacenter is less than its external costs, then it’s a net loss for society and ought to be shut down.