Paul Krugman writes:
Well, here’s the thing: regulation demonstrably does work where tort law doesn’t. Consider the environmental issue: in reality, the perpetrators of oil spills never pay most of the cost; but in reality, environmental regulation has led to much cleaner air and water. (Look up the history of Los Angeles smog or the fate of Lake Erie if you don’t believe me.)
So why does regulation work? If polluters can buy off the system ex post, after a disaster, why don’t they manage to totally corrupt regulation ex ante? There’s a lot to say about that, and I’m sure there’s a literature I haven’t read. But one thing we tend to forget in this age of Reagan is the importance and virtues of a dedicated bureaucracy: when you have professional government agencies with a job to do, and treat them with respect, that job often gets done.
I'll offer a few points:
1. Here is Susan Rose-Ackerman on regulation vs tort (JSTOR). This is a standard piece in the area, and although she does not stress public choice arguments, the upshot is not extremely far from Krugman's conclusions. She thinks that tort can replace regulation to only a limited degree.
2. I agree with Krugman that the "tort only" position espoused by some libertarians does not work. Yet I do not think his remarks are pointing at the best possible understanding of the question. For instance…
3. There is in fact an agency regulating off-shore drilling and in the case under question it totally failed. How can Lake Erie, an orthogonally related success, be cited but this very directly relevant failure not be mentioned?
Furthermore standard accounts of this failure blame regulatory capture, and the Congressional desire for revenue for this failure, not the ideology of free market economics. You might also blame voter sentiment, since Obama seemed to pick up some advocacy of off-shore drilling from the Republicans and arguably this happened because that position proved to be one of the few effective Republican arguments in the last election.
4. There are plenty of problems where tort works better than regulation, most notably when behavior cannot be controlled or measured easily ex ante or where the correct decision relies on the producer's decentralized information. This encompasses numerous areas, from medical practices to accident law to injuries resulting from the handling of guns to, most of all, many micro-components of highly regulated sectors. Here is a legal and moral defense of tort law.
5. Off-shore drilling is closely related to these cases. For a number of obvious reasons, it is hard for the regulator to monitor how safe an off-shore platform really is. We thus rely a lot on ex post penalties in this area (regulatory as well as tort), not because of ideology but because the ex ante proscriptions don't have as much bite as we would like. This reliance on the ex post penalty also means that "regulation plus tort" (never mind "regulation alone" or "tort alone") may not work so well in this area, with or without free market ideology.
6. Krugman seems to argue that regulators are more protected from the ravages of bad ideology than are judges. Maybe yes, maybe no (see below), but such a comparative argument is unnecessary. The most salient point for Krugman's conclusion is simply that optimality requires a bit of regulation and a bit of tort law and that to do without regulation is, in some regards, to be too lax. He could justify his main conclusion much more simply than he does. (That said, I still believe in less regulation than Krugman does, though I accept the logic of this argument.)
7. If we need to make the said comparison – the corruptibility of regulation vs. the corruptibility of tort law – an independent judiciary need not do worse than a respected bureaucracy and indeed the former is arguably more protected from political interference. It's also the case that the judiciary is more likely to overturn very bad behavior from other parts of the government. This is an advantage for judicial approaches over reguatory approaches and it is a further sign of (partial) judicial independence.
8. I worry when numerous bloggers and writers — not just Krugman — hold a primary theory of regulatory failure based on regulators who do not agree with them ideologically. This is at odds with a big chunk of the relevant public choice literature, which stresses knowledge and incentives, yet without ruling out ideological bias as a factor. Political science approaches to regulation offer greater scope for ideological bias as a major cause of regulatory failure, but still it is hardly the dominant theory to the exclusion of others. In this matter we should follow the literature, and evidence, which are there for the taking.