The State Department Final Supplemental Environmental Impact Statement (FSEIS) on the Keystone XL pipeline, published a year ago, concluded, reasonably, that the pipeline would have virtually no impact on global greenhouse gas (GHG) emissions or climate effects even under the most extreme assumptions. After all, the pipeline would transport 830,000 barrels per day of heavy oil-sands crude from Alberta to the United States; the annual GHG emissions from that oil on a life-cycle basis would be about 147-168 million metric tons. Under the assumption that no other oil production anywhere in the world would be displaced, that would add less than 0.4 percent to global GHG emissions of about 46 billion tons per year.

The attendant temperature effect in the year 2100 of those emissions under the highest IPCC assumption about the climate sensitivity of increasing GHG concentrations would be about four ten-thousandths of a degree Celsius. (The standard deviation of the temperature record is about eleven one-hundredths of a degree; accordingly, the impact of Keystone XL would be unmeasurable even apart from the fact that it would be effectively zero.) And even that abstracts from the reality that the oil will be produced and shipped with or without Keystone XL; indeed, rail and truck transport already move roughly 100,000 barrels of that oil per day, although far less safely.

Not so fast, says the EPA in a recent comment on the State Department analysis. Given the decline in world oil prices, Keystone XL, by reducing transportation costs, would increase the net price received by the Alberta producers, and so might increase oil sands output above levels likely to be observed at the lower world price, thus increasing GHG emissions and global climate impacts.

Now — to be blunt — this is nonsense on steroids. First: Keystone XL would reduce transport costs regardless of the level of international oil prices, and so in the EPA analytic framework would increase Alberta production whether prices are low or high. The EPA may have some sort of supply elasticity parameter in mind such that a price of, say, $50 per barrel would reduce production below levels that would support a pipeline project with a capacity of 830,000 barrels per day. (The State Department in chapter 1.4 of the FSEIS assumes a price range of $65-75 per barrel.) But why would TransCanada, the builder of the proposed pipeline, intentionally construct a pipeline with excess capacity if lower oil prices are expected to reduce the amount of oil sands production? Such excess capacity is expensive, and so the obvious answer is that they would not.

The reason that the pipeline will be built — and the new pipeline capacity will be used — even with oil prices substantially lower than those assumed by the State Department, is straightforward: Fixed (or sunk) costs for the Alberta oil are uniquely high relative to production costs compared with that relationship for conventional crude oil. The production cost (once the fixed costs are invested) of the existing Canadian oil sands projects is about $30-40 per barrel. Does EPA believe that prices will fall below that?

With Keystone XL, the Canadian oil would compete with Mexican and Venezuelan heavy crudes for utilization in Gulf Coast refineries, as those facilities are optimized for heavier crudes. Higher transport costs imposed by rail or truck shipments for the most part reduce net prices received by the Canadian producers. But that transport cost differential is not remotely sufficient to affect Alberta production; without Keystone XL the oil will continue to be produced, but refined in northern or eastern refineries, thus yielding an inefficient allocation of crude oil geographically, with lighter oil from hydraulic fracturing operations sent to the Gulf Coast refineries. That is why construction of Keystone XL would be likely to reduce gasoline prices on net, and that is why the Canadians, in the absence of Keystone XL, eventually will build their own pipeline, probably to the west, so as to ship the oil to Asia. Without the pipeline, increased demand for rail transport will continue to create railroad capacity constraints, and thus increased incentives to expand that capacity over time. The upshot: The Canadian oil will be produced, and the only questions are where it will be refined and at what higher cost. Over the longer term, investment in additional Canadian production capacity is likely to be reduced in the face of sharply lower world oil prices, but that has nothing to do with Keystone XL.

Second: The EPA comment argues in a nutshell that Keystone XL, by reducing transport costs, would offset to some substantial degree the effect of the recent decline in oil prices, thus raising oil sands output and GHG emissions above levels that would be observed without the pipeline. But nowhere does EPA offer an estimate of those effects.  Assume that without Keystone XL, lower oil prices would reduce oil sands output by half of the 830,000 barrels per day, and that construction of the pipeline would fully offset that effect. This means that the pipeline, under the EPA analysis, would increase global GHG emissions by 0.2 percent, and global temperatures in 2100 by two ten-thousandths of a degree.  Has anyone at EPA actually thought this issue through? Apparently not: The EPA comment letter does not offer any such estimates — they would be embarrassing — instead constructing an analogy between the pipeline and “5.7 million passenger vehicles or 7.8 coal fired power plants,” a blatant exercise in obfuscation rather than illumination.

The EPA comment letter illustrates well the larger reality that the bureaucracy — the regulatory bureaucracy in this case — is an interest group. Its analysis is driven by a particular preferred conclusion rather than the reverse. This is distinct from the effects of the pressures imposed by officials appointed by the president. Neither its analyses nor its pronouncements are to be taken at face value, as they are driven far more by ideology than by rigor.

Benjamin Zycher is the John G. Searle Chair and a resident scholar at the American Enterprise Institute (AEI), where he works on energy and environmental policy. He is also a senior fellow at the Pacific Research Institute.