One reason not to institute the same penalty for the police is that taxpayers end up paying. You might say, good, that means voters will try to ensure future p

One reason not to institute the same penalty for the police is that taxpayers end up paying. You might say, good, that means voters will try to ensure future police don't do this shit. Which is a nice theory, but probably doesn't work out in practice. In practice, you have small towns that can be bankrupted by a good-size suit, which is probably not a good outcome. You also have large cities that have room to minimax what they can get away with, and now all you've done is possibly deterred really egregious cases. It gets substantially more complicated with the intersection of feds and state agencies, where (depending on the state) police are good at gaming where the money goes depending on how and what is charged by whom. (There are complicated revenue sharing arrangements between, for instance, the DEA and local cops for forfeitures. If your hypothetical bank robber needs to be deprived of the money (and I don't think I agree, but that's a different discussion), the proper way to align incentives is to ensure the police do not see any of the money. Period. Give it back to the bank, put it in the state general fund, give it to a random orphan, burn it, whatever[1]. But as long as the police can by shiny toys with it (conference attendance, margarita machines and outright theft have all happened), they have an incentive to push it as hard as they can. [1] Giving the funds to the state still has incentive problems, but they are muted. I don't know that there's any empirical evidence if it is sufficiently muted to align police behavior with what most non-cops consider reasonable.