# Some Simple Arithmetic About Environmental Regulation and the Economy

Are environmental laws costing us jobs?  There are a number of reasons to question that idea, but the simplest one is simply that environmental law doesn’t cost enough to make huge economic effects plausible.

Let’s begin on the cost side.  To be fair to the anti-regulatory folks, I’m going to use the compliance cost estimate of a report from the Small Business Association of \$280 billion (2009 dollars).

The SBA has come in for a lot of criticism for its deregulatory slant, so if anything this number is probably too high.  (The report is from SBA’s “Office of Advocacy,” which tells you something about its objectivity.) A complete analysis would also have to take into account the benefits of the regulations (which include healthier and more productive workers), as well as job creation due to compliance activities.  But let’s forget all that for now.

\$280 billion seems like a lot of money. Actually, it really is a lot of money. But everything is relative: Compare that with the following numbers to see how small that is in terms of the economy as a whole:

GDP.  In 2009, GDP was about \$14,000 billion.  That’s fifty times the estimated cost of environmental compliance.  Also note that compliance costs in and of themselves don’t affect GDP — whatever the company is paying to comply is somebody’s income for the sale of labor, goods, or services, so those compliance activities get counted in GDP.  The effect on GDP has to be indirect, assuming there is one.

Corporate profits  In the third quarter of 2010 alone, corporate profits were \$1,659 billion (about twenty-three times the environmental compliance costs for that quarter using SBA’s estimate).  With this kind of money in corporate coffers, it seems unlikely that spending on environmental compliance is crowding out business investment in new plants or products to any significant effect.  (Not to mention the fact that corporations aren’t spending the money they have right now.)

Energy costs.  Total energy costs are about 7% of GDP, which would be about \$980 billion.   Even if energy costs are raised by environmental compliance costs, this would affect a relatively small piece of the economy (and even there it looks like the effect wouldn’t be huge.)

In short, on the face of things, environmental compliance costs seem to be too small to have much of an impact on the economy as a whole even if use a – estimate of the costs.  Maybe costs created by government have a multiplied effect on the economy — but if that’s true, the Defense Department budget must be a huge drag on the national economy, since it’s about 2 1/2 times bigger than SBA’s estimate of environmental compliance costs.

I’d like to contrast this lack of support for feedback effects with climate change.  CO2 is a trace gas in the atmosphere, and as you would expect, the direct effect of increases on temperatures is relatively small.  But we have very detailed models about how those effects get magnified, plus empirical evidence from ice cores and elsewhere tying together CO2 levels and global temperature.

We don’t have anything similar that would explain why environmental compliance costs would have an outsized effect on the economy out of proportion to the size of the costs (and presumably to the effect of other costs like defense spending).  We also don’t have any credible evidence that past changes in environmental costs have had a discernible effect on the economy as a whole.  So the idea of major hidden feedback effects lacks credibility.

Of course, all of this leaves out completely the fact that the regulations are saving lives, keeping fisheries safe from pollution, helping resorts make money by allowing people to swim safely, and so forth.  But you don’t even have to take that into account to be suspicious of the claim that environmental regulation is a huge drain on the economy.  In absolute terms, environmental compliance costs might be large, but relative to the economy as a whole, they’re not much more than a rounding error.