We have not added any noise to the clamor over EPA’s proposed rule under section 111(d) for existing power plants because it is neither the greatest thing since sliced bread nor does it spell the end of Western Civilization as we know it. After a month of reading and cogitation, we have the following four observations:
Economics 101. EPA proposes to reduce power plant CO2 emissions about 20% from 2012 levels by 2030, a reduction of about 1% a year, and a reduction in overall U.S. projected emissions in 2030 of about 8%. However, the EIA outlook for that period assumes only 4% cumulative increase in electricity demand between now and 2030.
It is certainly true that electricity demand growth has been minimal since 2008 (total end use is actually down 0.9% 2008-2012  and EIA’s 4% cumulative growth is based on the assumption of 2.4% GDP growth  combined with what EPA describes as structural changes in the economy favoring lower emitting businesses, and increased efficiency efforts. As for the latter, many would also argue that a major cause of the recent demand decrease is the loss of manufacturing jobs (the flip side of those “structural” changes) and above all, the slow growth of the economy since the recession.
Assuming a more traditional GDP growth of 2.8%, rather than 2.4%, EIA projects a 10% rise in electricity demand by 2030. The difference between a 2.4% and 2.8% GDP growth rate equals a 6% increase in electricity demand in 2030, which is a lot of capacity to find. In that case, we think some of the key components of EPA’s plan become problematic, e.g., it will be harder to keep the lights on while closing down old coal capacity; harder to improve the heat rate of the more modern coal plants because it will be more difficult to take them out of service for the retrofits needed; EPA’s assumed 11% demand reduction vs business as usual due to efficiency efforts – crucial to balance assumed supply – is jeopardized; and, last but not least, it will further fuel natural gas demand growth, making the economics of their assumed 70% gas dispatch goal more challenging.
Rube Goldberg. By going beyond the direct emissions reductions available “inside the fenceline” of these plants, the resulting proposal inevitably became astonishingly complex. Both EPA’s methodology for creating the individual state targets (comprising the four “building blocks” of direct emissions reductions, demand side efficiency, increased natural gas dispatch, and renewables) and what it expects to see in state implementation plans is opaque, to say the least. Hence the déjà vu: in this respect EPA’s proposal is the regulatory version of Waxman-Markey, an ominous precedent.
The silver lining to this complexity is that the task of trying to draft state implementation plans encompassing emissions reductions from improved heat rates at the EGUs, increased natural gas dispatch, demand-side efficiency and increased renewables, makes a state-level carbon tax more and more attractive. A transparent and predictable price on EGU CO2 emissions can achieve everything that those four building blocks can, at far lower cost to utilities, ratepayers, and the state. And the state gets to keep the money.
Crossing the Rubicon. In going “outside the fenceline”, EPA also crossed both a political and legal Rubicon.
Politics: Beyond the baseline political risk that was apparent even before the proposal came out – i.e., threatening states with the federal takeover of power generation if EPA was not satisfied with state implementation plans — the proposal reveals an entirely new risk as well. Measures to either create (or enhance) demand-side efficiency reductions, or implement (or increase) renewable energy standards (let alone creating new multi-state programs) will require, as EPA acknowledges, some, and probably a great deal, of state-level legislation. And even if the required cooperation were eventually forthcoming – a very, very big “if” – the delays inherent in state legislative processes mean that we need to add another couple of years to our previous 15-year generic 111(d) timeline .
The courts: The legal risks also cannot be overstated: The ink was not dry on the proposal before the first case was filed, claiming that EPA simply has no authority at all to regulate CO2 under Section 111(d). This claim is not specious: in fact, read literally, that is precisely what 111(d) says, and thus EPA will have to convince both the D.C. Circuit and the Supreme Court that it can not only impose this unprecedented regulatory regime, but can do so under a provision that on its face does not allow EPA to regulate CO2 even from the power plants themselves. Deference to EPA’s reading of the Clean Air Act is one thing; deference to an EPA interpretation in the service of outside-the-fence emissions reductions and threatened federal takeover of electricity production is another.
Moreover, EPA added to those legal risks when it simultaneously announced its proposed rule dealing with “modified” plants under section 111(b). While there is nothing notable about those standards (essentially that an existing plant that is modified remains subject to the existing plant standards) they spell trouble for this entire enterprise. Here is how:
The Clean Air Act requires that there be CO2 emission standards for new EGUs as a regulatory predicate for imposing such standards on existing EGUs. Assuming that EPA finalizes its proposed CCS standard for new plants, if the D.C. Circuit strikes down that standard, the existing plant standards are invalidated as well.
EPA understands that its proposed CCS standard is legally highly suspect. But rather than withdraw it in favor of a more defensible standard (IGCC, ultrasupercritical) EPA now claims that the modified plant standard could also serve as “the requisite predicate” for the existing plant rule. Unfortunately, EPA’s legal theory for how this works is even less likely to pass muster than the CCS standard itself, which leads us to conclude that the legal risks in both the new and existing plant rules means that it is better than 50-50 that the existing plant rules – certainly as proposed — will never come into effect.
Overall, we fear that political calculus behind both rules is ultimately based less on establishing a rational climate policy than on the PR benefits of giving the President something concrete to say at the Paris climate talks next year, and his desire to be able to claim a political legacy beyond that.
 We’re not sure why the sliced bread metaphor ever caught on, since anyone with a knife – a technology first employed by Homo habilis about 2.5 million years ago – can do so.
 EIA data at http://www.eia.gov/electricity/data.cfm#sales.
 EIA projection reference case at http://www.eia.gov/oiaf/aeo/tablebrowser/#release=AEO2014&subject=0-AEO2014&table=2-AEO2014®ion=1-0&cases=ref2014-d102413a.
 RIA for proposed rule, p. 88.
 EIA projection, op cit, high growth case.