4 ms·
Most theories of optimal fining are based on social costs S, discovery cost C, and discovery chance P. If, for example, you would like to maximize total welfare
by greenpresident 3y ago
Most theories of optimal fining are based on social costs S, discovery cost C, and discovery chance P. If, for example, you would like to maximize total welfare, you would fine S/P+C. A rational actor would only engage in the behavior of their expected gain was larger than the expected cost to society.
The firms profit should be irrelevant in my opinion. But make these firms believe that they will have to cover the true costs of their behavior!
- plagiarist 3y agoYeah, I mean, as long as we're idealizing, let's have them pay the actual costs that were externalized to society instead of merely wiping out profits from misdeeds.
- horsawlarway 3y ago> A rational actor would only engage in the behavior of their expected gain was larger than the expected cost to society. This is - bluntly put - a shite fucking idea. I am not trying to "maximize total welfare" where one way to accomplish that is to allow a bad actor to get ridiculously wealthy at the expense of everyone else (even if the total paid by everyone else is less than the bad actor makes). We have a term for this: Privatized profits and socialized losses. We're witnessing that this mindset is completely dysfunctional. It allows bad actors to accumulate immense wealth through behaviors that are net negatives for all other participants - further, those bad actors can then use that wealth to avoid or offset the share of their social costs that should be paid (ex: lobbying politicians, paying off regulators, bribing judges, etc) ---- There is a reason most people don't accept a deal where both parties gain, but one gains at a heavy advantage over the other: We know that allowing this behavior is bad for long term social stability, and we have all sorts of ideas about "Fair" that we apply to stop it: https://www.sciencedirect.com/topics/neuroscience/ultimatum-game https://www.sciencedirect.com/topics/neuroscience/ultimatum-... > The firms profit should be irrelevant in my opinion. But make these firms believe that they will have to cover the true costs of their behavior! The profit is the ONLY FUCKING THING that is relevant. The firms don't care about "true costs" (which, as an aside are often impossible to reasonably calculate). All the firms cares about is "Does bad behavior A lead to more net money, or less net money?" As long as the answer is "More net money" The firm is going to engage in bad behavior. The answer ALWAYS needs to be: Fine the bad behavior until it is not profitable. Full fucking stop.
- manoDev 3y agoIMO, all that is necessary is the law strongly backing personal liability of C-level in all industries. Trying to penalize companies never works in practice, regulators CANT let large companies going bankrupt, cascading defaults of creditors, creating unemployment, etc. It never happens. Also, this penalizes the workers the most. Once psychopaths have certainty of personal consequences they will want to stay away from executive positions.
- horsawlarway 3y agoThere is too strong an incentive to assign fall men to positions of "public responsibility" and then operate behind them at will (profitably). I'd maybe be in favor of a law that demanded that fines be paid in equity of the company (starting with equity held by the C-level officers and board) at a steep discount on market price (say... 50% valuation). Engage in enough bad behavior and you become nationalized over time, and then the public can decide what to do with the result.
- hedora 3y agoRegulators definitely could let large companies go bankrupt. They’d seize 100% of the stock and cancel any exit packages, then fire the board, and CEO, and claw back any remaining fines from the personal accounts and garnished wages of the executives responsible for the problem. This would protect the employees, customers and creditors of the company. Of course, it wouldn’t protect the people that own congress and the supreme court, so it will never happen.
- reaperman 3y agoTo be fair, if you follow the GP's formula, and magically plug in the correct numbers...you'll socialize the gains as well. Because the social losses of the pollution will be offset by the high fines. It's a fair system, it just moves agency for "who can pollute for the social good?" from the government to the private corporation. I don't love it personally, but it's decidedly not "Privatized profits and socialized losses". It's explicitly a system to make sure that the profits are shared from the damaging-corporation, to the government of the people damaged. In theory, I think they have a great point, which might be workable to inform future fine frameworks (we will always need fines in addition to anything else, and this isn't the worst way to set the fines). In reality, I suspect that captive government officials will set some of the variables far too "in-favor" of the corporations in exchange for backroom deals or cronyism. But it's far from the worst idea I've seen.
- hedora 3y agoThat is why punitive damages are typically 10x. (S/P+C is roughly 10). In this case, it’d be a 10B fine, which is 33% of their current market cap. That seems about right. They make other stuff, so letting the company survive (but barely) probably maximizes societal utility moving forward.