In his Tuesday night speech, President Donald Trump made reference, as he often does, to regulations that have killed American jobs.
This is an oft-used argument on the right — so common, in fact, that it is now taken as a kind of foundational truth, one that is simply self-evident, requiring no evidentiary support. It is one of the conservative economic catechisms (taxes slow growth, rich people create jobs, regulation kills jobs) that’s been repeated so frequently that even mainstream reporters tend to simply assume their truth.
But, at least in the case of the environmental regulations Trump is specifically attacking, it isn’t true. There is no consistent evidence that environmental regulations cause long-term changes in overall employment.
And in timely fashion, the Institute for Policy Integrity has a new brief with a clear and succinct explanation why this is so. Let’s walk through it.
The US economy is large. Regulations are, in comparison, small.
At the most basic level, the story is simple. Market economies tend to grow or contract based on broad macroeconomic factors such as aggregate demand, the rate of inflation, and population growth. As long as demand remains strong, inflation remains low, and the population grows, economic growth generally continues and employment rises. If those factors go the other way, growth slows and employment falls.
Within this broad framework, regional or sectoral developments (regulations on a particular industry, particular industries growing or declining, population shifts between and among regions) are generally lost in the national noise. If a particular resource declines, the market finds substitutes. If a particular industry declines, other industries grow and absorb the workers. If demographics shift, economic activity shifts with them.
This is, after all, the great brilliance of free market economies: They adapt. Massively parallel decision-making among millions of market actors turns out to be a very effective model for overcoming obstacles. Human ingenuity is boundless, etc. etc.
For some reason, when it comes to regulations, conservatives’ optimism and faith in markets vanish. If a scarcity or obstacle is artificially imposed by government through regulation (as opposed to happening “naturally” via historical circumstance or international competition), they rend their garments and lament that it will destroy jobs and growth.
But it isn’t true. The balance of evidence shows that environmental regulations have little to no effect on long-term aggregate employment. They may lead to reduced jobs in a particular sector, but as long as macroeconomic conditions are favorable, those jobs will simply move to other sectors. For instance, pollution standards might reduce employment in coal mining and coal-fired power, but demand for electricity will pull those jobs into either pollution remediation or alternative sources.
Of course it’s not just conservatives guilty of this; most political rhetoric about jobs is bogus, including much rhetoric from environmentalists about how various subsidies or tax credits create jobs. (For instance, there are many reasons to defend ARPA-E, but job creation isn’t one.) Such policies may boost jobs in a particular sector, but those effects will generally be balanced by losses elsewhere. Environmental regulations simply are not big enough to affect aggregate employment in a huge economy like America’s.
As this working paper from economists Marc Hafstead and Roberton Williams III concludes, “overall effects on employment are not a major issue for environmental policy.”
The funky, opaque models that support bogus job claims
Typically, jobs claims are backed by models, giving them a sheen of scientific plausibility, but the models economists use to estimate job effects have a number of important flaws. (IPI put out a longer explainer on this a few years ago, if you want to dig in.)
Model results are extremely sensitive to the assumptions that shape the model; small differences in those assumptions (which are often based on shortcuts and estimations rather than empirical research) can yield huge differences in results. Organizations that are, uh, invested in particular results can easily tweak assumptions to get them.
Those assumptions are often buried in the models and difficult to dig out and examine. (This is especially true with computable general equilibrium models, or CGEs.) That makes it difficult for laypeople to understand why different models yield different results — and difficult to compare or evaluate results.
Analyzing one sector of the economy for employment changes cannot, by definition, tell the full story of aggregate, economy-wide employment effects. (So conservatives can point to jobs lost in one area without noting jobs created elsewhere; environmentalists, vice versa.)
These battles between opaque models lead to absurd situations like this one, pulled from the IPI brief:
In one revealing example, the American Coalition for Clean Coal Electricity estimated that two EPA rules on power plant emissions would trigger a 1.4 million job loss; meanwhile, using a different model and different assumptions, the Political Economy Research Institute predicted the same two rules would generate a 1.4 million job gain.
This is flimflam in every direction. The truth is that the two EPA rules are likely to have small and short-term effects on total employment.
Local job effects matter! But so do public health benefits.
Just to be absolutely clear: The loss of jobs in local areas and local industries matters. People live in particular places and work in particular industries. They do not generally care about aggregate national employment; they care about their own lives, and so, rightly, do the politicians who represent them. Coal miners, for instance, are not likely to be comforted by the idea that a regulation that leads to loss in coal mining jobs will also lead to creation of wind manufacturing jobs.
There’s nothing deceptive or illegitimate about a politician opposing (or supporting) an environmental policy because of its local effects. All politics is local, as they say.
However! Since working-class employees of polluting industries are often cast as the victims of environmental policy, it’s also worth pointing out that the public health benefits of environmental regulations tend to wildly outweigh the employment and compliance costs.
An example from the brief:
EPA proposed controls for hazardous air pollutants, such as mercury, from industrial boilers in 2010. EPA estimated the rule would generate between $25.2 and $65.5 billion in annual net benefits, including up to 8,000 premature deaths avoided per year. By comparison, the agency estimated a cumulative, net employment effect on the regulated industry of between −4,000 and +8,300 jobs, with a central estimate of +2,100.
Even if the grimmest employment estimates play out in this case and 4,000 jobs are lost in particular industries, national legislators have to weigh those lost jobs against billions in public health benefits that will themselves disproportionately benefit lower-income and working-class people. It is owners, not workers, who most benefit from offloading health costs onto the public.
There’s a weird background assumption in these discussions that economic/employment benefits and losses are real, while public health is a “social” benefit, a kind of affective bonus that makes lefties feel good.
But health is very much an economic issue, and public health benefits are very real. Supporting or opposing a regulation purely on the basis of its local job effects is simply not good policy; those effects should be balanced against the regulation’s benefits.
And if local employment is a priority, there are complementary policies (economic development, job training, etc.) that can address those effects. Denying tens of thousands of people better health is a poor way of supporting local jobs.
Better debate, please
It’s no great revelation to say that political debates tend to be somewhat dumbed down and involve a lot of sloganeering on all sides. That’s certainly true of environmental policy debates, which very rarely involve a nuanced discussion of trade-offs.
“Regulations kill jobs” is mostly just sloganeering. Environmental regulations can often move jobs from one industry to another or one region to another. They often have short-term employment effects. But by and large, they simply don’t have any substantial long-term effects on US employment.
A smarter debate would focus on their local or industry-specific job effects and weigh them against the public health benefits they produce. That kind of inclusive cost-benefit analysis often supports environmental regulations, though, so demagoguery on jobs is unlikely to go anywhere.