IER’s Robert Murphy on the Social Cost of Carbon

Last week the Senate Environment and Public Works Committee (EPW) held a hearing titled “Climate Change: It’s Happening Now.” That’s right folks, global warming is not going to strike “The Day After Tomorrow,” as alarmists previously predicted. It’s going to happen “Two Days Before the Day After Tomorrow” — today. 

But before you sell the beach house, move to North Dakota, or join a survivalist group, you might want to read the testimonies by University of Colorado Prof. Roger Pielke, Jr., University of Alabama in Huntsville Prof. Roy Spencer, and Institute for Energy Research scholar Robert Murphy.

I’ll discuss Murphy’s testimony today and Pielke, Jr.’s later this week. (I covered the basic argument of Roy Spencer’s testimony in a previous post: Climate Models: “Epic Failure” or “Spot on Consistent” with Observed Warming?)

Murphy challenges the intellectual bona fides of the Obama administration’s May 2013 Technical Support Document (TSD) on the social cost of carbon (SCC). Climate activists increasingly invoke SCC estimates to justify the imposition of carbon taxes, fuel economy mandates, Soviet-style production quota for wind farms, fracking bans, and other interventions to rig the marketplace against reliable, affordable, fossil energy. They speak as if SCC estimates disclose an objective reality like the boiling point of water or the specific gravity of iron. In fact, SCC estimates are assumption-driven hocus-pocus or, as my colleague Myron Ebell prefers to say, “hogwash.”

SCC analysts purport to measure the damage, in monetary terms, that an incremental ton of carbon dioxide (CO2) emissions inflicts on humanity and the biosphere. As discussed previously on this blog, SCC estimates depend on assumptions about highly speculative issues such as climate sensitivity (how feedback mechanisms, positive or negative, will amplify or damp down the direct warming effect of rising greenhouse gas concentrations), climate impacts (how projected warming will affect weather patterns, ice-sheet dynamics, and eco-system services), economic impacts (how projected changes in global temperature, weather, and sea-level rise will affect agriculture, forestry, tourism, and other climate-related activities), and technological change (how adaptive capacities will develop as climate changes).

Each layer of the analysis is fraught with uncertainty and is educated guesswork at best. By adjusting the assumptions, the SCC analyst can get pretty much any result he desires.

Murphy zeroes in on the simplest part of the analysis: Which discount rates federal agencies use to estimate the present value of future projected climate change damages.  

Discounting is a necessary feature of cost-benefit analysis, especially for regulations designed to address climate change, in which most of the damages are assumed to occur decades or even centuries into the future.

People typically value costs and benefits in the present more highly than those in the future. For example, if the minimal compensation you will accept to do a job is $50 paid up front, you will not do the job for $50 paid a year from now, and certainly not for $50 paid a hundred years from now. The more distant and uncertain any future gain or loss, the more its present value declines. The rate at which the present value of a future cost or benefit declines is the discount rate. (Note: Discount rates differ from person to person and even for the same person under different circumstances.)

SCC estimates critically depend on the choice of discount rates. Set the discount rate very high — in other words, assume that people today attach little value to costs or benefits incurred 50-100 years hence — and the SCC becomes vanishingly small. Conversely, set the discount rate very low — in other words, assume that people today care a great deal about costs or benefits incurred long after their lifetimes — and the SCC becomes very large. Unsurprisingly, the selection of discount rates in SCC analysis is a bone of contention.

With that as background let’s turn to Murphy’s testimony.

Murphy contends that in the May 2013 TSD, the Obama administration’s Interagency Working Group on the Social Cost of Carbon flouts Office of Management and Budget (OMB) best practices in regulatory accounting. OMB Circular A-4 instructs agencies to use a 7% discount rate as the base case in regulatory analysis, because that is the “average before-tax rate of return to private capital” in the U.S. economy. “Yet even though the guidance from OMB was quite explicit on this point,” Murphy writes, “both the initial White House Working Group report from 2010, as well as the recent update in May, did not report the SCC using a 7 percent discount rate; they only used discount rates  of 2.5, 3, and 5 percent.”

That’s odd since the OMB is a member of the Interagency Working Group. Maybe they’re too busy saving the planet to follow their own guidance.

How significant is this omission? In the May 2013 TSD, the SCC for 2010 is $11 per ton at a 5% discount rate but $52 per ton at a 2.5% discount rate. “In other words,” Murphy comments, “cutting the discount rate in half caused the reported SCC to more than quadruple.” Clearly, the choice of discount rate heavily influences SCC estimation. What would the SCC be with a 7% discount rate? Murphy opines:

If the Working Group ran the computer models again, this time using a 7 percent discount rate and an earlier reference year such as 2015, presumably a larger fraction of simulations would register zero or negative values for the SCC, so that the mean result would itself be closer to zero—or conceivably even negative, meaning that carbon dioxide emissions conferred extra benefits on humanity.

In another trick contrary to OMB guidance, the Working Group calculated the global SCC but not the domestic SCC. The 2010 TSD acknowledges that, “Under current OMB guidance contained in Circular A-4, analysis of economically significant proposed and final regulations from the domestic perspective is required, while analysis from the international perspective is optional” (p. 10). Yet, notes Murphy, the May 2013 update reports only global SCC estimates. The effect is to make climate change appear to be a worse problem for the U.S. than the underlying analysis actually indicates.

The global SCC, after all, incorporates SCC estimates for developing countries, which have fewer resources for adapting to climate change. How much does the U.S. domestic SCC differ from the global SCC? According to the 2010 TSD, “a range of values from 7 to 23 percent should be used to adjust the global SCC to calculate domestic effects.” In other words, if, using a 3% discount rate, the global SCC in 2030 is $33 per ton, the corresponding domestic impact is only $2-8 per ton.

As with omitting SCC estimates using a 7% discount rate, reporting only global SCC estimates and omitting domestic estimates creates a pro-regulation bias. Murphy explains:

When the May 2013 update came out, the headline media reports typically focused on the SCC figure for the year 2010 at a 3 percent discount rate, which was $33/ton; this value was often reported as “the” social cost of carbon. Yet this was a  global estimate of the SCC. If instead the default reports were expressed from the domestic perspective, then the same 2010 figure at a 3 percent discount rate would only have been in the range of $2 to $8 per ton.

To see the significance of this decision by the Working Group, consider the following scenario: Suppose the EPA issues a new regulation that causes private industry to restrict carbon emissions, and that the compliance costs (in terms of forfeited economic output in the U.S. because of the new regulation) work out to $25/ton. Using the Working Group’s recent headline SCC estimate of $33/ton, this regulation would apparently pass a cost/benefit test, because the $25 cost to American industry for every ton of restricted emissions would be counterbalanced by $33 in avoided future climate change damage. However, Americans would still on net be hurt by the regulation, as they would only receive $2 to $8 of the stipulated benefits (i.e. avoiding the domestic social cost of carbon on each ton no longer emitted), while suffering the full $25 in compliance costs.

The only point I would add to Murphy’s trenchant analysis is that even if (a) the climatological, meteorological, and technological forecasts underpinning SCC estimates were accurate, (b) agencies use appropriate discount rates, and (c) agencies use U.S. domestic SCC estimates in cost-benefit analysis, SCC analyses would still be one-sided and misleading.

Even if scrupulously based on the best science and economics, SCC analyses would still ignore the social benefits — the positive externalities — of affordable, reliable, carbon-based energy. Consequently, such analyses turn a blind eye to the social costs — the adverse effects on public health and welfare — of the economic losses imposed by carbon mitigation schemes.

Unless paired with a rigorous and thorough analysis of climate policy risk, SCC analyses are functionally biased and partisan. Will we ever get a fair and balanced assessment from SCC analysts? Fat chance.

One thing we can be certain of in this universe of flux and change: The Obama administration will never convene an interagency working group on the social cost of carbon mitigation.