The Most Important Part of the Keystone XL Environmental Impact Statement

The State Department has released its long-awaited final environmental impact statement (EIS) for the Keystone XL pipeline. The headline is straightforward: the pipeline is “unlikely to significantly impact the rate of extraction in the oil sands” and, as a result, world greenhouse gas emissions. This is essentially a status quo conclusion, reaffirming the essence of the draft EIS (released last year). It also allows President Obama to judge that the pipeline meets his requirement that the project “not significantly exacerbate the problem of climate pollution”. The report does, however, carve out one substantial exception. That’s worth drilling down into, because it’s what the President will likely lean on if he decides to say no.

The logic in the final EIS (as in the draft) is straightforward: blocking the Keystone XL pipeline is unlikely to significantly affect oil sands production because oil sands has other ways of getting to markets. But the final EIS, unlike the draft one, stress tests that claim, pushing it to identify conditions under which it would fail. It finds some, but they’re narrow.

The first condition for Keystone to have a significant impact on oil sands extraction has to do with other pipelines: they need to consistently fail for Keystone to matter. Otherwise, denial of Keystone would shift oil sands to different routes that have similar economics, with roughly the same emissions results as Keystone itself.

The second condition has to do with rail. If no pipelines are built, oil sands will be shipped by rail instead, with producers incurring higher transport costs as a result. The draft EIS noted this, but argued that oil prices were so high relative to the breakeven price for new oil sands investment that producers would essentially eat the extra transport costs without cutting production. Producers’ profits would drop, but extraction would remain the same, as would emissions.

The final EIS leans on this with a simple question: is there any oil price at which this argument would fail? Its answer is yes – which is at the heart of the exception that the EIS carves out. At an oil price between $65 and $75, it says, the extra cost of transporting oil by rail rather than pipeline would flip some producers’ economics from the black into the red, prompting them to leave some Alberta oil in the ground. (Below $65, the rail versus pipeline distinction would again be mostly moot, since oil sands extraction would quickly become uneconomic in general.) Here’s what the EIS says:

“Assuming prices fell in this range, higher transportation costs could have a substantial impact on oil sands production levels—possibly in excess of the capacity of the proposed Project—because many in situ projects are estimated to break even around these levels. Prices below this range would challenge the supply costs of many projects, regardless of pipeline constraints, but higher transport costs could further curtail production.”

The EIS authors note that a $65-$75 oil price is below most long-run estimates. As a result they treat this possibility as an outlier.

My guess is that their worst-case scenario is even less likely than they think.

Why?

Because it’s difficult to conjure a low oil price case in which oil sands production is sharply constrained. To the extent that some forecasters expect oil prices to fall and stay below $75, it’s because they see a surge of non-OPEC oil supplies combining with low demand to push prices down. A significant contributor to that surge, of course, is new output from the Canadian oil sands. The upshot is that if forecasters were asked for projections in which oil sands were trapped in Alberta, even fewer would project oil prices below $75 than do today.

Here’s another way to look at this. Imagine that oil prices go down to $70 because of a mix of supply gains and demand curbs -- the sort of scenario the EIS flags. And now imagine that, because of constrained transportation options, the average break-even cost of a new oil sands development suddenly rises from $65 to $75, blocking some oil sands investment. What happens to oil prices? They go up, either to squeeze demand or to boost supplies, in order to make up for the missing Canadian oil. Those higher prices, in turn, weigh against the higher transport costs, bringing some Canadian projects back into the black.

Among other things, this makes it difficult to believe the EIS claim that blocking oil sands pipelines could ultimately have an impact “possibly in excess of the capacity of the proposed Project”. There are self-correcting mechanisms – less oil production in one place means higher oil prices everywhere and consequently more oil output somewhere else – that would weigh strongly against this.

To be certain, even if blocking the pipeline were to keep Canadian oil in the ground, that wouldn’t make killing Keystone a good idea. Just because there could be climate impacts from the pipeline (and, to be clear, I think there would be some small ones) doesn’t mean that they would be large. More important, any decision on the pipeline will need to go well beyond climate and take economics and international relations into central consideration as well. If the President decides to reject Keystone, though, it will be on climate grounds. And the final EIS shows that, if he wants to do that, he’ll need to thread a very small needle.