Scott Pruitt, Donald Trump’s nominee to lead the Environmental Protection Agency, has raised more hackles among progressive Democrats than any other Trump cabinet nominee. Typical of the ferocious opposition to his candidacy is the screed prepared by the Sierra Club that deems him a mortal threat to the safety of the planet because, as Attorney General in Oklahoma, he has “spent his time in office working to allow big polluters to do whatever they want, rather than protecting the health, clean air and water of his constituents.” Democrats like Senator Brian Schatz of Hawaii have insisted that his nomination is “a four-alarm fire” because Pruitt is a pawn of fossil fuel companies whose cardinal sin is denying the conclusion of “climate scientists” that human emission of carbon dioxide is creating a global warming crisis.
The defenders of Pruitt have been equally vocal. President Trump, no man to mince words, has railed against the EPA for spending “taxpayer dollars on an out-of-control anti-energy agenda that has destroyed millions of jobs, while also undermining our incredible farmers and many other businesses and industries at every turn.” In his view, Pruitt is needed to restore some sense of balance to the entire enterprise.
So the big question is who is right and why. In order to cut through the strong rhetoric on both sides, it is important to return to first principles to see how the regulation of pollution and jobs fit together. To start, neither the environmentalists nor Pruitt hold the untenable view that pollution is a good thing. But they profoundly disagree on two key second-order questions—the correct definition of pollution and the right remedial structure to deal with its occurrence.
Many of Pruitt’s attackers appeal to the narrow definition of pollution that includes harms everyone accepts—emissions into the air, water, and ground of toxic substances that cause discomfort, disease, and property destruction. Unfortunately, however, the mission of the EPA as they see it goes far beyond the sensible objective of addressing these problems. Today’s fight against carbon dioxide emissions has nothing to do with health effects like asthma, mercury poisoning, or premature deaths. Rather, the dispute is chiefly over the adverse effects of carbon dioxide on global warming. On that issue, Pruitt is right to insist that uncertainties on the social cost of carbon are, as Judith Curry writes, uncertain. There are first differences in the estimate of the sensitivity of the changes in carbon dioxide on temperature, which seem lower today than, say, a decade below. But there is massive ignorance of how to project the social and economic consequences of these changes over a period of 300 years. The estimates really matter because overstating the risk could shut down activities, like much fossil fuel use, that are net positives to the overall economy.
Global warming is not the only area in which modern environmentalists overshoot the mark. They also use a definition of environmental harm that cuts too broadly in, for example, the destruction of habitat needed for wildlife preservation. Notwithstanding its surface plausibility, this definition unwisely over-expands the regulatory arm of the state. Thus, in the 1995 Supreme Court decision Babbitt v. Sweet Home Chapter of Oregon, the Supreme Court upheld regulations from the Department of the Interior that treated habitat destruction as the equivalent of killing or maiming animals.
Sweet Home missed the strong differences between the two cases. Killing animals has always been a civil wrong, and the government may use coercion to prevent the wanton destruction of animals to preserve the common stock. But imposing habitat protection cuts a far deeper swath into gainful private activities when a simple state designation, especially in the case of birds, can turn vast acreage into development-free zones. No private party could impose these massive restrictions on the use of private property without paying for the privilege. The EPA may have the power to do so unilaterally, but it should do so only if it pays. This would remove the differential burdens on landowners, which, as Gary Libecap recently noted, are thrown on those landowners in the path of government regulation and not on the public at large. Given this unfortunate turn, the EPA faces no price constraint preventing it from gobbling up resources with multiple uses. Landowners, however, well understand that even if habitat is a public good, it is also a private bad, so they have every incentive to eliminate it before the government comes along. If compensation were required they would preserve and promote their habitat in order to be able to sell it to the government.
Nonetheless, in yet another unfortunate application of Chevron deference to the EPA, Justice Stevens, writing for the Supreme Court majority, lets the government decide whether to proceed by using its power of eminent domain or by regulating without compensation. No government agency should receive such a wide berth, because it is virtually a foregone conclusion that regulation will typically displace compensation, thereby unduly expanding state power. In all those cases, it is easy to demand that everyone desist. It would be a transactional nightmare for the government to try to buy out all of the thousands of potential hunters to preserve wildlife. It is of course easy for the government to negotiate the purchase of needed habitat. Conservation groups make these purchases all the time.
The current law is equally defective in its choice of remedies in the event of pollution. Everyone agrees that polluters should ordinarily be required to pay for the damage they cause to both public and private property, as was long required under the common law. But one key element in the private law equation was to wait until the potential nuisance was imminent or actual before issuing an injunction. The EPA does not worry about these limitations in the exercise of its enormous permit power, but requires the proponents of any new project to run a huge regulatory gauntlet that consumes years and many millions of dollars before anything can be done.
For example, in the waning days of the Obama administration, the Department of the Army, under strong pressure from the Standing Rock Sioux Tribe, required Dakota Access to prepare an exhaustive Environmental Impact Statement to complete its 1,100 mile pipeline, without once mentioning where the previous studies had been inadequate. Decisions like this can cause social losses in the billions of dollars, for which the government bears no financial responsibility at all. And worse still, that particular action increased environmental risk by increasing oil shipments by rail and truck, which are far less controlled environments.
Yet another risk element in environmental policy is granting subsidies to one class of players over another as part of a general system of industrial policy. Just that happened with the billions in cash subsidies for wind and solar power, including those for failed projects like Solyndra. The situation is even starker when they receive additional in-kind subsidies in the form of exemptions from various laws, including the Endangered Species Act. Subsidies always distort optimal resource use, and environmental politics are no exception to the general rule. The correct approach in all cases is to tax and regulate those parties that create nuisances, at which point there is no reason to give separate subsidies to their competitors. Overregulation is always a risk that matters.
The current set of policies introduces high levels of economic waste and distortion, wholly without reference to jobs. The question is whether President Trump’s claim about the loss of millions of jobs adds anything to the discussion. One plausible response is that Trump is looking at only half the picture, jobs lost through regulation, while ignoring the flip side, jobs created through and by these same regulations. Just that position was taken recently by Alana Semuels writing in The Atlantic, who stresses two points—first, that regulation necessarily kills jobs to make air and water cleaner and safer; and second, that the loss of jobs in one sector is offset by the creation of jobs in another sector, so that the aggregate effect is often a wash.
Unfortunately, a whiff of misguided Keynesian economics pervades that analysis. The mistaken premise treats jobs solely as a benefit, when they are also a cost, which diverts resources from one area to another. The hard question is whether the net benefits generated by these jobs justifies the costs that are imposed. To see why, suppose someone could build the Dakota Access pipeline for half its cost without compromising quality. It would be foolish to say that the greater project efficiencies were job-killing changes. Instead, the more accurate accounting sees the same asset, the pipeline, built for less, and the release of additional labor to perform other tasks that in turn can produce either new capital assets or consumption benefits. It is only in a topsy-turvy world that these savings should be regarded as job-killing. It is for just that reason that the once-common union practice of featherbedding in the railroad industry, by keeping firemen fully employed in non-work on diesel trains, was rightly regarded as a social waste.
Under the implicit Keynesian logic, it should make no difference if we employ one group of workers to dig a hole, and a second group to fill it up again. That practice creates two jobs. But it is far different from a system in which those two sets of workers each drill one hole worth more than it costs, for now at the end of the day the two sets of wage payments produce two capital assets instead of none.
Just this problem arises in the environmental area. It is appropriate that jobs be lost if they are the source of pollution, given the benefits that are secured elsewhere. But there are no offsetting benefits when vital resources are used for excessive permitting that generates no benefits at all. Nor is it ever appropriate to suggest, as Semuels does, that tough regulation is fine so long as it is imposed with equal severity “because everyone is experiencing the same rise in costs.” That proposition is wrong for two reasons. First, it is rarely the case that uniform regulation will have the same effect on cost for all firms within an industry. There is typically some level of disparate impact where the firms in question use different techniques of production. It is for just this reason that taxes tend to outperform specific regulations—they allow each firm to seek its own best method of compliance. Second, even if the impact were uniform, it could well be that the regulations are so severe across the board that the entire industry shuts down under its weight.
The key point then is to design environmental controls with an eye to stopping the external harms. Once those calculations are made, firms within the industry should have, at least in some cases, an incentive to alter their modes of production to comply with those standards. At the same time, the economy as a whole benefits from the improved overall environment.
At this point, therefore, we should reframe the Trump indictment of current environmental practices to read that misguided regulation causes a loss in overall social output that in turn eliminates useful jobs that would survive under a sounder regulatory regime. No one could argue that the resources spent on compliance should be set at zero. But by the same token, environmental regulatory reform is desperately needed to correct major flaws in the current institutional design. Let’s hope that when the shouting dies down, Pruitt is up to the task.