In theory, yes you’re right there might be a perverse incentive.
But in practice I don’t think it’ll be very pronounced, because regulators do not have that much influence on a corporation.
Also the individuals regulator’s salaries are not directly tied to stock dividends of the corporations they regulate, but rather the absolute amount of staff in a team tasked with regulating a particular corporation or sector within a larger government agency.
On the individual level promotions, pay rises, bonuses, and other incentives should still be based on what we’re ultimately want regulators to do: reducing negative externalities and increasing positive externalities (i.e. performance reviews, doing good work).
Adding more checks and balances (e.g. journalism, watchdogs, critical think tanks, red teaming within government) might be better than what we have now.
I agree that those incentives should be based on what we want regulators to do. And I don’t think the salaries would be explicitly tied to stock dividends of the corporations they regulate. But the profits of those corporations would affect the budgets of the regulators, which seems like it could in practice influence salaries, incentives, etc. If we’re already facing a notable amount of regulatory capture, then clearly things like oversight are not fully keeping regulators aligned with what we want them to do. And my guess, though again it’s not based on much, is that the effect of “intensifying perverse incentives for regulators, as long as no one’s looking too closely” would be greater than the effect of “sometimes providing additional resources for regulators”.
I guess another reason that might be so is that the potential for perverse incentives could always be there, whereas the additional resources only kick in when a company actually does provide more dividends, which is probably correlated pretty well but not perfectly with power, which is correlated pretty well but not perfectly with negative externalities.
And another might be that it seems like limited resources is just one problem, and I’m not sure it’s the main one. My very vague impression is that the concern is often more about the corruption or capture of regulators. So it seems like that proposed intervention might target the smaller cause of the problem, while potentially exacerbating the larger cause.
Thanks for the comments… I think you’re generally raising a bunch of valid points.
A few more thoughts on this:
First, perverse incentives are already present within the current system. Indeed, they’re likely ubiquitous and inherent in any optimization process—it’s just a matter of how powerful they are and the scale.
There is already a lot of regulatory capture and “revolving doors” between regulators and industry (c.f. 2008). In theory, nuclear regulators should want the nuclear industry to flourish, because their job and job prospects depend on it. In practice, it seems hard for me to see how implementing this would increase perverse incentives. Again, if you do this on an industry wide basis (the department of nuclear security is funded by an ETF of nuclear companies), it seems hard to see how regulators can substantially affect the stock movements of a whole industry.
It seems like a related but slightly separate problem.
Also note that this strategy is neutral wrt the absolute level of funding of the regulators. Rather it helps with prioritization within regulation (i.e. the government can buy $10bn or $100bn in stocks and disburse them to fund regulators).
Second, generally, I slightly disagree that limited resources for regulators is not one of the major problem (again see 2008 - where regulators are always way behind industry in terms of staff, resources, analytic capacity, etc. - the same seems to be true for the tech industry, where there aren’t sufficient funds to hire top talent with deep understanding of the sector). Again, this strategy doesn’t really speak to the absolute amount we should spend on regulation—just that it’s a better way to cut the pie.
Hm… but I actually just had another idea of how to fund regulators perhaps reducing perverse incentives and also optimizing more for negative externalities.
Consider that the EU has fined Google $10bn. I think currently this is disbursed to go into the general budget.
But maybe one could buy Google stocks (or a technology ETF) with that money and use the dividends to fund digital regulation.
This way regulators would be incentivized to reduce negative externalities (through fining companies and the industry), but then also they’d be held back to completely wreck industries or companies, because they’re financed by the overall health of the industry after the fine.
In theory, yes you’re right there might be a perverse incentive.
But in practice I don’t think it’ll be very pronounced, because regulators do not have that much influence on a corporation.
Also the individuals regulator’s salaries are not directly tied to stock dividends of the corporations they regulate, but rather the absolute amount of staff in a team tasked with regulating a particular corporation or sector within a larger government agency.
On the individual level promotions, pay rises, bonuses, and other incentives should still be based on what we’re ultimately want regulators to do: reducing negative externalities and increasing positive externalities (i.e. performance reviews, doing good work).
Adding more checks and balances (e.g. journalism, watchdogs, critical think tanks, red teaming within government) might be better than what we have now.
I agree that those incentives should be based on what we want regulators to do. And I don’t think the salaries would be explicitly tied to stock dividends of the corporations they regulate. But the profits of those corporations would affect the budgets of the regulators, which seems like it could in practice influence salaries, incentives, etc. If we’re already facing a notable amount of regulatory capture, then clearly things like oversight are not fully keeping regulators aligned with what we want them to do. And my guess, though again it’s not based on much, is that the effect of “intensifying perverse incentives for regulators, as long as no one’s looking too closely” would be greater than the effect of “sometimes providing additional resources for regulators”.
I guess another reason that might be so is that the potential for perverse incentives could always be there, whereas the additional resources only kick in when a company actually does provide more dividends, which is probably correlated pretty well but not perfectly with power, which is correlated pretty well but not perfectly with negative externalities.
And another might be that it seems like limited resources is just one problem, and I’m not sure it’s the main one. My very vague impression is that the concern is often more about the corruption or capture of regulators. So it seems like that proposed intervention might target the smaller cause of the problem, while potentially exacerbating the larger cause.
Thanks for the comments… I think you’re generally raising a bunch of valid points.
A few more thoughts on this:
First, perverse incentives are already present within the current system. Indeed, they’re likely ubiquitous and inherent in any optimization process—it’s just a matter of how powerful they are and the scale.
There is already a lot of regulatory capture and “revolving doors” between regulators and industry (c.f. 2008). In theory, nuclear regulators should want the nuclear industry to flourish, because their job and job prospects depend on it. In practice, it seems hard for me to see how implementing this would increase perverse incentives. Again, if you do this on an industry wide basis (the department of nuclear security is funded by an ETF of nuclear companies), it seems hard to see how regulators can substantially affect the stock movements of a whole industry.
It seems like a related but slightly separate problem.
Also note that this strategy is neutral wrt the absolute level of funding of the regulators. Rather it helps with prioritization within regulation (i.e. the government can buy $10bn or $100bn in stocks and disburse them to fund regulators).
And then it can’t do all the prioritization without human input: perhaps we do not need to regulate the $4 trillion global restaurant industry as much as the $4 trillion dollar global chemical industry. But I think there’s currently not much quantitative prioritization going on wrt how much regulation capacity should be assigned to different companies within an industry. So this is just to finesse current funding levels.
Second, generally, I slightly disagree that limited resources for regulators is not one of the major problem (again see 2008 - where regulators are always way behind industry in terms of staff, resources, analytic capacity, etc. - the same seems to be true for the tech industry, where there aren’t sufficient funds to hire top talent with deep understanding of the sector). Again, this strategy doesn’t really speak to the absolute amount we should spend on regulation—just that it’s a better way to cut the pie.
Hm… but I actually just had another idea of how to fund regulators perhaps reducing perverse incentives and also optimizing more for negative externalities.
Consider that the EU has fined Google $10bn. I think currently this is disbursed to go into the general budget.
But maybe one could buy Google stocks (or a technology ETF) with that money and use the dividends to fund digital regulation.
This way regulators would be incentivized to reduce negative externalities (through fining companies and the industry), but then also they’d be held back to completely wreck industries or companies, because they’re financed by the overall health of the industry after the fine.