The Obama Administration has released its 645-page proposal that mandates existing U.S. power plants reduce their greenhouse gas emissions 30 percent below 2005 levels by the year 2030. Experts knew that such a rule was coming, and last week the U.S. Chamber of Commerce released a study outlining the huge costs such a measure would pose on the U.S. economy—including an average of $50 billion in forfeited GDP every year through 2030.

Ironically, some of the Administration’s most loyal partisans didn’t dispute the Chamber’s numbers. Instead, they claimed its impacts (which also include a $586 billion cumulative loss in disposable income for Americans through 2030) were cheap. Paul Krugman wrote that the Chamber study proved that Obama’s power plant rule “would have surprisingly small economic costs.” So right off the bat, it’s important to note that fans of the regulation do not dispute the Chamber’s cost estimate of $50 billion in GDP per year.

Now of course, we can quibble about what constitutes “surprisingly small economic costs”; $50 billion a year in lost output might seem like chump change to Dr. Krugman, whereas it might seem more expensive to a struggling business owner or household. But a separate point is that the Chamber study shows that the power plant regulations fail the Obama Administration’s own cost/benefit test for climate policy. It’s not even close. This whole episode thus underscores that nobody in Washington actually takes these “social cost of carbon” numbers seriously; they ignore them when they give a failing grade to a proposal they want to push through.

The Chamber Study versus Administration’s “Social Cost of Carbon” Estimates

The crucial rhetorical trick here is that supporters of the power plant regulation focus on the total price tag, and utterly ignore what Americans supposedly get (in the form of reduced climate change)for this sacrifice. Imagine a wife looking over the monthly credit card statement and seeing that her husband has been going out to get sushi at lunch several times a week. When she complains that their food budget is really tight, the husband shoots back, “Hey! I only spend $60 a week on sushi. You spend way more than that at the grocery store. So you see, my food policy is incredibly cheap.”

That’s what’s going on here, with the Administration’s power plant regulations. For example, Krugman doesn’t actually quantify the benefits, in order to compare them with the quantified costs. Instead he just writes, “The Chamber’s supposed scare headline is that regulations would cost the US economy $50.2 billion per year…That’s for a plan to reduce GHG emissions 40 percent from their 2005 level, so it’s for real action.”

Does that sound like a proper cost/benefit analysis? In any other context, would it sound rigorous for a Nobel Prize-winning economist to contrast $50.2 billion in annual costs with benefits merely described as “real action”?

Fortunately for us, the Chamber study did actually quantify the ostensible benefits of their modeled rule, in terms of avoided tons of carbon dioxide equivalent emissions. Here is what the Chamber concluded (see pages 7-8 of the study):

The economic cost to achieve each ton of emissions reduction also is extraordinarily high. This analysis indicates that the additional cuts in CO2 emissions in the Policy Case come with an average price tag of $51 billion per year in lost GDP over the forecast period, which translates into an average undiscounted economic cost of $143 per ton of CO2 reduced. When EIA modeled the Waxman-Markey cap-and-trade bill, the economic cost per ton of CO2 in its “Basic” scenario averaged an undiscounted $82 over the same period, still quite high but considerably less than the $143 figure arrived at under the Policy Case [for the modeled power plant regulations].

The economic cost for each ton of reduced CO2 in the Policy Case also exceeds the upwardly revised social cost of carbon (SCC) estimates developed by the Administration’s Interagency Working Group on Social Cost of Carbon in 2013…. [T]he Working Group estimated that by 2030, the SCC will have risen to between $17 and $82 per ton (in 2012 dollars). Applying the same range of discount rates, the average cost in the Policy Case ranges from $153 to $163 per ton over the analysis period, much higher than even the Working Group’s 2030 figure.[U.S. Chamber study, pp. 7-8, bold added.]

The above statements are bombshells, so let’s make sure the reader understands their significance. Emissions-reduction regulations imposed just on power plants—as opposed to the entire country—are of course going to have lower total economic impacts, because the range of their harm is not as large as a more sweeping cap-and-trade bill that applied to the entire country. But by the very same logic, the ostensible benefits of a regulation that just hits power plants (rather than the whole country) will also be limited, because the reduction in emissions is only applicable to power plants, rather than automobiles and other businesses, etc.

Therefore, any decent economist knows that to evaluate whether the regulation makes economic sense, you compare the costs with the benefits. The Chamber study looked at a hypothetical rule (based on their information at the time) that would reduce power plant emissions by 40 percent through 2030,[1] and found that the cost per ton of avoided emissions would be $153 to $163, depending on the discount rate used.

So is this a good deal? Does it make sense to forfeit about $158 in economic output, for every ton of reduced emissions?

No, it doesn’t, according to the Administration’s own Working Group on the Social Cost of Carbon. Their own computer models and parameter choices led them to announce back in May 2013 that the “social cost of carbon” in the year 2030 would range from $17 to $82 per ton (in 2012 dollars).[2] Thus, if Krugman and other defenders of the Administration were intellectually honest, they would have to tell their readers, “The Chamber study shows that power plant regulations fail a cost/benefit test by about a 3-to-1 margin.”


The U.S. Chamber study found that regulations imposing power plant emissions reductions would have enormous absolute impacts, to the tune of $50 billion in lost GDP per year. The supporters of the regulation (such as Paul Krugman) didn’t dispute the Chamber’s numbers; instead they tried to defend them as being “surprisingly small.” If we move beyond the absolute impact, and look instead at the cost relative to the ostensible climate benefit then things get even worse; on this score, the policy studied by the Chamber would have a relative cost almost 75% more than Waxman-Markey.

To add to the irony, the Chamber’s analysis shows that power plant regulations would fail the Obama Administration’s own cost/benefit criterion for climate regulations. The whole point of setting up a Working Group to issue estimates of the “social cost of carbon”—as dubious as this process is, as we’ve demonstrated here at IER—was to give federal agencies guidance when running cost/benefit analyses on their proposed regulations. Yet as this episode shows, nobody in Washington actually takes these SCC numbers seriously, if they get in the way of a pet policy.


[1] To be clear, the Obama Administration’s actual rule requires a 30 percent reduction by 2030. This information had not been finalized when the Chamber commissioned its study.

[2] The actual numbers reported in the 2013 Working Group technical update show 2030 SCC figures of $16 through $76 per ton, depending on the discount rate. However, those figures are 2007 dollars, while the Chamber analysis was in 2012 dollars.