Notes on Today’s Existing Source Rule
Today, EPA announced proposed performance standards for existing power plants under the Clean Air Act (also known as the existing source performance standards, or ESPS). Attention to the proposal is deservedly high, and it’s a monster of a rule (645 pages, at least in this unformatted version). That means general overviews are already out, and in-depth ones will have to wait. But there is still plenty to say today. Last week, I listed the five things I was most interested in seeing in the rule. Here’s how they came out.
First, this post will be about some in-the-weeds details. That’s likely what you come here for, so I doubt that’s a problem. But it’s worth taking a step back and appreciating that this proposal, if it survives political and legal challenge, will be the single most significant action on climate the U.S. government has ever undertaken. It won’t be successful unless it can move other countries to take action, but it has a better chance of doing that than any speech or vague commitment. It’s far from perfect in its current form, and will remain flawed in important ways, but it gets a lot of emissions reductions at reasonable cost, and it’s what we’ve got. Someday soon I hope we can replace it with something even better.
Second, a point on structure. The proposal would set a separate goal for each state based on a formula that takes into account the various emissions-cutting opportunities available in that state’s power sector. This is similar to an approach taken in NRDC’s shadow proposal from last year (so some credit goes to them for either prophecy or influence). There are pros and cons to this strategy, particularly in terms of implications for states that are big importers or exporters of power. It will be interesting to see what energy economists and engineers predict.
The big headline is 30% emissions cuts from the power sector relative to 2005 by 2030. That’s quite large – over 500 million tons of CO2 per year. Most of those reductions (25%) appear to come by 2020, though a closer reading reveals states have some flexibility to push compliance out toward 2030 so long as they show they are making progress. That explains why, as Michael Levi observes, the 2030 target is not much tougher than the 2020 target. Think of it as a “2020s” target and a final push goal in 2030. I suspect that EPA expects these targets will be superseded by 2030, either by a new rule or, I think much more likely, by legislation (presumably setting a carbon price).
Also, remember that emissions have declined since 2005 – a good chunk of that 30% is already baked in.
The proposal doesn’t say a lot about flexibility, but that’s probably a good thing. Most of the proposal is dedicated to explaining how EPA will set targets for states to meet in their plans, and to the procedure for submitting, reviewing, and updating those plans. Very little is devoted to explaining the substance of the plans – that is, what states can actually do achieve the goals. But this lack of detail gives freedom to states. And EPA does often suggest in the proposal that states may consider market-based approaches, and actively encourages states to band together and create regional programs by giving participating states more time to submit their plans. Carbon taxes, however, are not mentioned in the rule. Carbon-tax backers have argued that states will be unwilling to consider that option unless EPA were to explicitly bless it. We may see whether that prediction is correct.
3) Coal and Gas
The proposal does envision shifting from coal to gas – probably the lowest-cost emissions reduction opportunity available in the US economy (assuming, of course, that fugitive methane emissions from gas operations are not too high). In setting states’ targets, EPA assumes existing gas plants can be run more (up to 70% of capacity), displacing coal. States aren’t required to force this outcome, of course, but they can do so – and can do much more, including setting up trading programs between gas and coal. Two important caveats, however:
a) EPA still left ambiguous the question of whether coal and gas would be grouped into the same “source category” for regulatory purposes – it “co-proposed” the existing separate categories and a new, combined category, just as it did in the proposed new source standards in January. My view is that combined categories are probably necessary for coal-gas trading.
b) EPA doesn’t consider the opportunity to cut emissions by building new gas plants to replace existing coal – just running existing plants more. Building new plants is could be a bigger opportunity, and there’s good evidence that it’s also cheaper (though EPA disagrees). Just because EPA doesn’t consider this opportunity when setting targets doesn’t mean states can’t when meeting those targets in their plans – in fact EPA explicitly says states can do so. But this might not be easy. Sources built after January of this year are treated as new sources outside the scope of the ESPS, and are never brought in. That makes it difficult, maybe impossible to bring them into a trading program – in essence, new gas becomes an “outside the fence” option just like renewables or energy efficiency.
In setting targets, EPA is considering efficiency improvements at existing coal plants and switching dispatch to existing gas plants as mentioned above. It’s also proposing to count new nuclear, renewables (as required by state portfolio standards), and a wedge of demand-side energy efficiency. States are free to push all of these and other opportunities for emissions cuts from the sector. Offsets and anything else that cuts CO2 outside the power sector are not considered. As I said last week, I think this set of options is legal, though I worry about the practical problems of verification and attribution with renewables and energy efficiency. EPA appears to assume, for example, that energy efficiency investments directly replace fossil generation. That may not be true – in a state with a renewable portfolio standard, we know it isn’t.
EPA tries to get around the attribution problem by proposing that states, not the regulated power plants, bear legal responsibility for renewables and energy efficiency policies. That’s a clever move, but my first impression is that its not legal. The statute requires state plans to set performance standards “for” emissions sources. EPA wants a broad reading of “for” that allows most any policy aimed at helping reduce those sources’ emissions to be part of the state plans. I don’t think that fits with either the ordinary meaning of “for” or the source-focused structure of §111. It also puts EPA in the position of dictating state energy policies that become enforceable under federal law (though EPA attempts to dodge that last outcome with further creativity).
To be fair to EPA, their creativity here is praiseworthy – they’re trying to avoid practical problems with policies aimed at reducing the cost of compliance. And they visibly struggle with the legal issues here in the proposal, articulating the legal arguments against their positions and requesting comment. Moreover, even if standards are enforceable only against power plants (not states or anyone else), that doesn’t completely rule out energy efficiency and renewables as compliance options. It just makes them a lot harder to do.
5) Mass or Rate
The state-by-state targets EPA proposes setting are emissions rates. There’s even a table in the proposal showing the proposed targets for each state. Many of them are quite low, under 500 tons of CO2 per megawatt hour when a new, efficient gas plant might emit 800. But renewables, nuclear, and energy efficiency are included in the “adjusted” numbers in the table – neither the state fossil power sector nor any individual fossil plant would be expected to meet these standards.
The targets vary greatly – from 264 lbs CO2/MWh (Washington) to 1,882 (Montana). I worry that this will cause perverse outcomes, wide disparity in compliance costs, or both. But I’ll withhold judgment until modeling comes out. Certainly EPA’s intent in setting independent state standards is to minimize these effects, not exacerbate them.
States are permitted to convert these emissions rate targets into mass targets (i.e. caps). I haven’t been able to find anywhere in the proposal that details how this calculation must be done, so it may be left up to states to explain their methodology in their plans (or I may simply have missed it). In any case, I’m also worried about perverse outcomes if states in the same power market or emissions trading program choose mass or rate standards.
Look for more on the proposal from us at RFF in the near future.