In the first and second parts of this series, we took a deeper dive into EPA’s new and improved Base Case projection accompanying the Clean Power Plan (CPP) Final Rule. Specifically, we looked at what the new projections held in store for coal-fired electricity generation and compared it to what EPA was projecting in its Proposed Rule. The changes are quite remarkable.
Part II of our analysis ended by noting how implausible it was that the power sector will achieve carbon dioxide emission cuts equivalent to 50% of those needed to meet the president’s 32% reduction target in 2030 compared to the 2005 level.
But let’s leave aside the leaps of faith needed to accept this and address how EPA expects to get the other 50%.
The administration brags that its plan will drive greater investment and commercial use of renewable technologies. Does it? Not really.
EPA’s own model run of the CPP Final Rule shows very little in the way of additional renewable generation beyond what’s already in its new and improved Base Case. By 2030, EPA estimates that CPP will produce just a 4% increase in renewable generation. In fact, in earlier years, EPA shows renewable generation falling below the Base Case. (Also curious is EPA’s assumption that these renewables facilities will produce about a quarter more energy than history suggests, something my colleague Dan Byers will address in a future “Fuzzy Math” post.)
And what are we to make of EPA assertions that, “Under the Clean Power Plan, by 2030, renewables will account for 28 percent of our capacity, up from 22 percent in the proposed rule”? What EPA fails to mention is that under the revised Base Case it trotted out for the Final Rule, renewables are slated to comprise 25% of generating capacity in 2030 even if CPP is not implemented. So half of the rise in total renewable capacity from 22% to 28% has nothing to do with CPP and everything to do with EPA’s finagling with the Base Case.
But if the additional emissions cuts aren’t coming from greatly increased use of renewables, how does the agency expect to get them?
By clobbering electricity demand growth and—what else?—shutting down coal plants. In our earlier posts detailing EPA’s revised business-as-usual scenario, we saw how the agency assumes that essentially every watt hour of generation lost from coal plants, about 200 billion watt-hours worth, would be replaced by a watt hour from a renewable plant. In the scenario is produced for the Final Rule, however, EPA assumes further losses in coal-plant output will be replaced by “nega-watts,” or energy efficiency.
EPA’s model run of the Final Rule adds another 320 billion watt hours to the decline in coal plant output already noted in its Base Case, an amount that matches almost exactly the total decline in electricity generation of 355 billion watt hours forecast in 2030 (with the difference being made up largely out of lower generation from natural gas plants).
In this regard, it’s worth pointing out that in its Final Rule EPA abandoned its proposed Best System of Emission Reduction (BSER) Building Block 4, the one devoted to improving energy efficiency and conservation through demand-side management. Despite the absence of Building Block 4, EPA’s still banks on much lower demand to drive even more coal plants into retirement and, more importantly, to rein in costs.
The Energy Information Administration’s (EIA) takes a different view of the potential contribution of energy efficiency. Although EIA’s modeling analysis of EPA’s Proposed Rule—which included Building Block 4—came up with even greater emission reductions than EPA estimates for its Final Rule (about 34% below the 2005 level in 2030 versus EPA’s 32%), EIA’s analysis relied much less on declining electricity demand and more on increased output from renewables.
When looking for emission reductions, both EIA’s and EPA’s models are geared to choose the least expensive compliance option first, and this gives energy efficiency the inside track. That EIA’s achieved even greater emission reductions than EPA without much of a contribution from energy efficiency indicates that EIA takes a more hard-headed viewed of the potential of demand-side management to deliver emission reductions. In short, EIA doesn’t buy EPA’s overly-optimistic claims about energy efficiency.
Why is all this important? Because as we documented earlier, EPA conveniently pushed a lot of the expensive stuff into its revised Base Case model run, which means these costs are not attributable to CPP. What we’re seeing in this instance is EPA sticking with unrealistic assumptions on energy efficiency in its CPP model run—even though demand-side management is no longer a central plank of its BSER scheme. Those unrealistic assumptions allow the agency to claim that compliance costs of the Final Rule are much lower than they would be with a more credible estimate of energy efficiency in line with what EIA produced.
This is what’s known as rigging the game. When combined, these flaws produce cost estimates for CPP that are unrealistically low.
It’s bad enough that EPA is stretching its legal authority beyond the bounds of the Clean Air Act. (Who could have guessed that an obscure 300-word section of the Act could cause such mischief?) But by low-balling its cost estimates, EPA is attempting to lull states in believing that they can completely overhaul their electricity systems for a pittance and with no impacts on reliability.
If something seems too good to be true, it probably is. Especially when it’s coming from EPA.