Blog Post

The Regulatory Limiting Condition

Jun 11, 2012 | Nathan Richardson

Marginal Revolution's Alex Tabbarok argues that "our regulatory system has gotten so large and complex that it’s main effects are now unintended. In short, the product of the regulatory system is a result of human action but not of human design."

This seems right to me, though it's impossible to prove (and Tabbarok doesn't claim it is).

The context is the observation that disruptive innovations (automated cars and tacocopters are given as examples) face greater legal and regulatory problems than technical or financial hurdles. Those legal barriers exist not by design but as unintended consequences of the regulatory system, which is fundamentally static or at least slow-moving, and was designed with past (not present, much less future) circumstances in mind.

This is true for environmental regulation too, but the lessons are mixed. Like any regulation, environmental rules have unintended consequences. Sometimes these can be bad - for example, power plants may install devices to reduce sulfur dioxide (SO2) emissions that also reduce their efficiency, and therefore increase their carbon emissions. But unintended consequences can also be good - reducing SO2 emissions also reduces airborne particulate matter (PM) since the two are chemically related. The negative health effects of PM weren't well understood when SO2 limits were set by Congress in 1990, but have turned out to be a huge part of that regulation's public health benefits.

These "co-benefits" can lead to contorted policymaking - the EPA now uses PM health benefits to justify deeper cuts in SO2 emissions, and new regulation on mercury and other "toxic" pollution. Some have criticized this as improper, and from a philosophical (and possibly legal) point of view they have a point. But the benefits are there and appear to exceed the known costs of the regulation.

This illustrates another feature that is important for Tabbarok's point about regulatory change - the contrast between flexible and inflexible regulation. The original SO2 limits were fixed by Congress. But for subsequent rules the EPA has used its broad Clean Air Act powers to protect public health and welfare from pollution. When new scientific information is available, EPA can (indeed must) update its rules. Congress, on the other hand, has barely touched air pollution regulation since 1990.

Which model you prefer depends, of course, on whether you trust regulators to get things right. But the legislative route is so slow that it exacerbates the problems Tabbarok points out. If technocrats can't keep up with innovation, politicians never will.

So what's the prescription? The unstated moral of Tabbarok's story is that we should be willing to change or scrap regulation that has outlived its usefulness. This is as true (maybe more so) for environmental regulation as for other parts of the legal system. But we rarely do this until things get very bad, and there aren't good tools for it. Rules get socially enshrined and attract their own constituencies, creating a big status quo bias. Cost-benefit analysis is generally done only when rules are first implemented and not reassessed later when real information is available (though the President has asked agencies to change this practice). Conversely, it's easy for regulatory reform efforts to get hijacked or demonized.

I suspect Tabbarok believes that the social and legal background can never change fast enough to keep up with technology, so they'll be the limiting condition on growth for the foreseeable future. But if that's true, the reaction shouldn't be to give up, but rather to focus on that limiting condition. Maybe we need some regulatory R&D.