The fine shouldn't be that low. Here's the best formula I could figure out. Let's say R1 = actual risk of harm to yourself and others, M1 = magnitude of harm t

The fine shouldn't be that low. Here's the best formula I could figure out. Let's say R1 = actual risk of harm to yourself and others, M1 = magnitude of harm to yourself and others, R2 = your perceived risk of harm to yourself, M2 = magnitude of harm to yourself. Both M1 and M2 should be based on market price of QALY (about 50K dollars?) without overvaluing you specifically. Then the fine should be set at R1⋅M1-R2⋅M2. That's a generalization of Pigovian tax to a spectrum of behaviors: pollution (R1=R2, M1>M2), skipping medical checkups (R1>R2, M1=M2), speeding (R1>R2, M1>M2). Does that make sense?