Environmental Regulation and Jobs, Again

Over at Econbrowser, James Hamilton argues that environmental regulation may be a job-killer after all.  There are two themes: the first is that US trade is weighted toward natural resources, and regulation is raising costs and reducing capacity in these tradables, and the second is that, in recessionary times, jobs lost due to regulations are not regained elsewhere.  I think he overstates his points, but he clarifies important issues that tend to get muddied in economic debates.


Lets take the second first.  I had argued that regulations tend to lower the measured productivity of workers in regulated industries, leading to some combination of more employment to restore output and expenditure-switching, as consumers shift to different products.  Hamilton’s examples—California agriculture, Texas oil and lignite, Alabama cement kilns—do not dissuade me.  We will continue to eat, power and pave, and if we do a bit less of some of this (especially the powering and paving), we can shift to more benign activities.

Hamilton is certainly right that it is important to observe gross and not only net labor flows into and out of employment; this has played a revolutionary role in our understanding the microeconomics of aggregate labor markets—search theory and all that.  To assume, however, that increases in separations resulting from regulation will not be offset by increases in hiring is simply to assume that expenditure-switching does not take place.  Does it?

Here is an example from my part of the world (at the moment).  Back in 2000, the Social Democrat/Green coalition government of Germany announced plans to shut down that country’s nuclear power industry by 2020.  When the Christian Democrat/Free Democrat coalition took over later in the decade, however, they backed off from this timetable.  This sparked a wave of protests, heightened by the disaster at Fukushima.  Now the CDU/FDP government has doubled back and endorsed a 2022 cutoff date.

Will this cause a loss of jobs?  No doubt workers in the nuclear power sector will suffer, and they may be unable to find new work that pays as well as the old.  But the nuclear phaseout is a boon to the offshore wind energy sector, where an influx of new investment is accelerating plans to turn the German Bight into a giant wind farm.  (Building a forest of wind turbines, intersected by a few shipping lanes, will bring about a transformation of the marine environment as profound and unprecedented as any human beings have attempted, but that is a story for another day.)  This is already leading to a hiring spurt, visible, for instance, in the town of Norden.  In other words, expenditure switching can actually happen—although, in this case it is also assisted with a large dollop of industrial policy, yet another story for yet another day.

Hamilton’s first argument, that environmental regulations tend to target natural resources, which underpin the US trade position, is paradoxical, but you have to know the history of this issue to understand why.  Beginning in the 1980s, environmentalists began to tell horror stories about how regulations were driving industries to third world export platforms—pollution havens—and that the fear of this effect was undermining regulation itself.  You will find most of these examples in the popular press and in political broadsides of groups like Greenpeace and the Sierra Club, although a few made it into the journals.  (See especially the work of the late Duane Chapman of Cornell, such as this.)

The economics profession responded to this “race to the bottom” claim by subjecting it to cross-section analysis: do industries with tighter environmental regulations tend to migrate to less-regulated regions, ceteris paribus?  The result has been a bevy of studies, whose general conclusion is that this has not been the case, or at least not to any great extent.  I’ll take this opportunity to state that I am not convinced by this work, for three reasons:

1. These studies do not usually control for shipping costs as a proportion of value added.  Extractables have the biggest environmental costs, but they are also (with some exceptions) heavy and expensive to transport around the world.  Failure to include this variable biases the estimated effect of regulation.

2. We do not have good direct measures of regulatory intensity, especially at the retail, enforcement level, and it is common to use proxies like pollution control expenditures.  This conflates the full cost of environmental mitigation, however, with the incremental effect of regulation.  Firms control pollution for a variety of reasons—potential tort liability, risk of damage to their equipment and operations, the occasional urge to do the right thing—and regulation is just one aspect.  Whether the proxies really proxy what they purport is an open question, in my opinion.

3. Recall that the hypothesis in question is whether trade competition leads to a race to the bottom in environmental regulation.  If it is true, it means that there will tend to be less regulation in contexts where trade impacts are envisioned.  In other words, causation is suspected to run from trade exposure to regulation as well as the other way around.  Models that treat regulation as exogenous, as all of them do, are therefore ruling out the hypothesis by assumption.

These are reasons to suspect that we still know very little about the race-to-the-bottom hypothesis.  On the other hand, I don’t think we can simply generalize from the examples picked by advocates; some sort of aggregate analysis is needed.  At the moment, all we can say is that the hypothesis seems to work in some cases, but its scope and scale are unknown.

Meanwhile, what is paradoxical about Hamilton’s post is that one of our best econometricians is going to battle against the econometric literature on the pollution haven effect by citing examples of specific regulations that he thinks will dampen our resource-based trade advantages.  He’s using Greenpeace methodology, but with priorities reversed.

Finally, to tie up a couple of loose ends, mention needs to be made about offsetting mechanisms.  Suppose there really is a pollution haven effect, and the US loses jobs because it increases regulation in tradables.  Are there any processes that can spring into action to restore employment?

One, much beloved by trade theorists, is the doctrine of comparative advantage.  According to this principle, regulation cannot alter the balance of trade, but only its composition.  If we export less of this, we will import less of something else, so the new jobs will come in some other trade-competing industry.  The problem, alas, is that the assumption that trade balances are unaffected at the margin (that they are determined solely by aggregate savings behavior) is just that, an assumption, and one that is contradicted by the weight of empirical evidence.  That’s why it is reasonable to worry about races to the bottom in environment, labor, taxation, and other realms.

The other is the potential for monetary and fiscal policy to counteract adverse trade effects: if we lose jobs on the trade front, why not amp up macropolicy correspondingly?  This is a complex issue, but, in the interest of bringing to a close a post that has already gone on much too long, let’s just say that expansionary policies are not completely free lunches, nor do they always work as advertised, nor is the political path to their realization unstrewn with obstacles.  It’s not wise to ignore the employment effects of microeconomic policies under the assumption that they can be fixed at the macro level.  But it is also true that macro policies can and should be adjusted in tandem with micro choices.