On March 27, 2012, the US Environmental Protection Agency proposed a New Source Performance Standards (under sec. 111 of the Clean Air Act) that would, for the first time, limit carbon emissions from power plants. I've finally had a chance to read through all 257 pages of the new rule (here) and another 100+ pages of EPA's Regulatory Impact Analysis (RIA) for the new rule (here).
The proposal would affect new fossil-fuel-fired power plants with generating units producing at least 25 megawatts of electricity, limiting their CO2 emissions to a maximum of 1,000 pounds per megawatt-hour. Currently, the average coal-fired plant in the US emits 1,600 pounds, while the average natural gas-fired plant emits 800 pounds (see here). So, under the new standard, natural gas-fired plants would not need to make any changes - indeed, 95% of such plants built since 2005 already meet the standard. However, electric plants powered by coal or coke are going to have to make changes to reduce emissions, including (EPA presumes) the introduction of carbon capture and storage (CCS) technologies, which remove emissions before they leave the smokestack and transport them to an underground storage facility with appropriate rock formations to hold the emissions safely (without substantial leaks) for decades or centuries. The graphic below indicates how the CCS process works (in theory):
EPA claims that its new rule will not impose major new costs on utilities or raise electricity prices for consumers based on economic projections showing that the vast majority of new power plants built over the next decade are expected to be gas-fired, rather than coal-fired (based on projected fuel prices), and so would meet the standard even without the rule. (Obviously, if it were the case that virtually all new power plants would meet the standard in the absence of the rule, the rule itself would be otiose.) EPA projects that by 2020 an additional 27 gigawatts of power will be added to the US electricity grid, of which only 2 gigawatts expected to come from coal, compared to 10 gigawatts from natural gas (with the rest coming from a variety of other fuel sources, including renewables, hydropower, and nuclear).
Even assuming (optimistically) that EPA's economic projections are correct, the agency makes an even more dubious technological assumption that "CCS is expected to become more widely available, which should lead to lower costs and improved performance over time" (see here). As quoted, that assumption is almost unassailable simply because EPA fails to provide any assessment of when we can expect that improved performance and lower costs - 5 years, 10 years, 100 years? A 2009 article in The Economist (see here) explained that, despite its technological promise, CCS remains unproven in practice, and transporting and storing carbon emissions, then monitoring for potential leaks over very long periods of time, could prove very costly, at least in the short-run (say, the next 10 years). In addition, the institutional framework needed to support CCS (including potential property-rights in underground storage facilities and liability for leaks) remains substantially under-developed (despite some recent EPA rules designed to facilitate CCS, see here and here).
EPA's RIA for the new rule argues, in accordance with the dubious assumptions stated above, that it is not a "major" new rule subject to cost-benefit analysis requirements because it is not expected to have an annual effect on the national economy of $100 million or more. Needless to say, some (but not all) power companies have offered a very different assessment of the social costs of the new rule (see here), which is not to say that they are right and EPA is wrong. However, EPA's heavy reliance on questionable assumptions (especially the technological assumptions about CCS) and lack of any real cost-benefit analysis for the new rule are hardly reassuring. (Part of the problem here, of course, is that it would be very difficult for EPA to assess the benefits of the new rule given the nature of CO2 as a highly diffuse pollutant and the problem of estimating net avoided climate change by policies, like this one, that rely on substitution of natural gas for coal.)
Don't get me wrong: I'm glad to see EPA finally tackling greenhouse gas emissions from the largest sources of those emissions. And EPA could justify the regulation's expectations for CCS as a "technology-forcing," in the same way that (ultimately successful) auto emissions limitations adopted in the 1970s could not be met by any technologies existing at the time the rules were promulgated (see, e.g., here). It's just that the Clean Air Act is a relatively poor vehicle for implementing a climate policy. Unfortunately, Congress's failure to enact a legal regime better suited to greenhouse gases, has left the federal government with no better alternative.