We all know that yelling “fire” in a crowded theater is an abuse of free speech. Anyone doing that yelling will be called to account for their behavior. We should expect the same in the NERC portrayals of the EPA’s Clean Power Plan. It’s time to see if there is evidence in their analyses that supports their alarms—and I don’t see how they justify the calls for delays.
NERC released an Initial Assessment of the EPA draft 111(d) rules on CO2 on April 21, but did not make any of its data and assumptions available until Friday afternoon, May 1. The renewable energy section of the inputs was still withheld at that time, and only released May 4.
First number to consider: 13
13 is the number of days between the release of the report and the release of the inputs and assumptions.
Key numbers for Clean Power Plan
NERC withheld its inputs and assumptions for 13 days after the release of its Clean Power Plan report. Why? Perhaps because their case is not helped by looking at the numbers. Photo: startswithabang.com
NERC’s alarm over plant retirements due to the EPA’s Clean Power Plan, especially coal plants retirements, doesn’t have the numbers behind it. Older, inefficient coal plants are already retiring, due to a variety of factors including low natural gas prices, the need to install pollution controls to meet public health standards, reductions in electricity demand, and the falling costs of renewable energy. NERC acknowledges this transition away from coal and to gas and renewable energy, but rather than address these changes in its recommendations, NERC emphasizes the EPA rule out of proportion to its impact.
NERC and others have projected power supply economics and concluded that just using market prices, old power plants adding up to something like 80-90 gigawatts (GW) are likely to retire by 2030. But half of these retirements have already been announced and are projected to occur by 2016, the deadline for complying with Mercury and Air Toxics rules. NERC estimates plant retirements at “approximately 43 GW” total due to the Clean Power Plan (NERC report, p. 34).
Here’s where the numbers get interesting. For the two regions where NERC projects a shortage of generation, NERC’s analysis shows that the EPA’s Clean Power Plan rule is to blame for closing 2 percent of the generating capacity in Texas, and less than 1 percent in the New York-New England region. The power industry in these areas are very sensitive to market competition. NERC’s analysis shows the retirement of old plants is happening regardless of the CPP, and will continue through 2030. In the much larger Midwest region, states provide revenues for utility-owned plants, so any forecast of closing based on economics is incomplete without the state 111(d) implementation plans.
Cause for alarm? The replacements are already expected!
The executive summary reports the projected closings, and then waits for the next page to mention the gas-fired replacement plants expected throughout the U.S. by 2020 add up to 60 GW and “approximately 80 GW by 2030.” NERC listed 78 GW of specific planned gas-fired power plants in the assumptions that NERC “forced” into the model. That is, to make the model reflect NERC’s expectations of the future, 78 GW of replacement capacity was built in. These planned units are included with greater certainty than NERC and its consultant EVA gave to planned renewable energy additions.
What about renewable energy?
NERC has more to explain about its renewable energy numbers.
There is no list of planned additions for renewable energy. Instead, NERC and its consultant developed a table of upper limits of how much renewables are allowed. The report hints that EVA employed an internally developed state-by-state forecast of renewable capacity. (See more in this EVA report for coal producer Peabody Energy.) That separate forecast of renewable capacity appeared as a table of input assumptions on May 4.
NERC-EVA explain that the renewable energy additions were limited by four factors, but “state renewable resource quality and potential” and “historical/forecast future renewable generation growth trends” (NERC report, p. 8) could be used to create any number from 1 to 1,000.
The number is zero
The table of inputs shows how much wind or solar might be allowed state-by-state in NERC’s modeling: Alabama, Mississippi, and Louisiana are shown with no solar in the future, South Carolina is limited to adding 1 MW between 2020 and 2030. Arizona, 40 – 50 MW of solar per year. New Hampshire, North Carolina, and Virginia: no wind. In very successful windy states, Kansas is limited to 25 MW, Oklahoma 600 MW, and Nebraska 62 MW total new wind between 2020 and 2030. NERC says plainly “based on NERC’s modeling assumptions, renewable resources primarily [are] being built to accommodate renewable portfolio standards.” (NERC report, p. 13)
The impact of these fixed input assumptions cannot be dismissed. NERC’s central argument is the power industry needs more time to build clean energy facilities, but has limited wind and solar energy contributions, in some cases to zero. NERC’s own survey included in this report shows wind and solar facilities have a 2-year shorter plan/build timeline than gas plants.
How bad are the results?
The results are surprising. NERC-EVA used two different models to project future impacts of the Clean Power Plan on the electricity sector. In results from the IPM model, no new wind can be expected for Kansas, Oklahoma, Nebraska or rest of the Southwest Power Pool (SPP), and none in Texas. In the other model, the results show wind added at the rate of 200 MW per year in the ERCOT part of Texas and 100 MW per year in all the SPP.
This flies in the face of recent trends and plans for future development. For example, Texas had over 8,000 MW and SPP had over 3,175 MW of wind under construction or under development with a power purchase agreement at the end of 2014, according to AWEA. AWEA also reports 95.8 GW of wind projects were in transmission queues around the U.S. at the end of 2014, including 23.9 GW in ERCOT and 14.3 GW in SPP.
NERC-EVA defies credibility with projections of zero new wind added from 2016 onward in these regions where wind performs better and cheaper than anywhere in the U.S. In 2013, average wind power prices in the interior region of the country were $22/MWh with the production tax credit, according to definitive energy lab study. Even without the PTC, projects in this part of the country would be cost-competitive with new natural gas combined cycle plants.
NERC’s projections for renewables are also pessimistic at the national level. NERC’s IPM modeling projections of total generation from renewables with the EPA rule in place in 2020 and 2030 is slightly lower than what EIA projects will occur in their Annual Energy Outlook 2014 reference case, not including the Clean Power Plan. NERC’s IPM modeling shows that non-hydro renewables would only increase from approximately 7 percent of total U.S. electricity generation in 2014 to 9 percent in 2020 and 10 percent in 2030.
Compare with others’ numbers
Recent analyses by UCS and the U.S. Department of Energy show that the U.S. can achieve much higher levels of renewables with little to no impact on electricity prices.
A UCS analysis completed last fall using NREL’s ReEDs model and more realistic assumptions for renewables found that EPA could nearly double the amount of renewable energy included, to 22 percent of total U.S. electricity generation by 2030, with a 0.3 percent increase in consumer electricity prices and a 9 percent reduction in natural gas prices by 2030. This would also reduce total power sector CO2 emission reductions from 30 percent below 2005 levels by 2030 to approximately 40 percent.
These results are consistent with the recent DOE Wind Study, which found that generating 20 percent of U.S. electricity with wind power by 2030 would increase consumer electricity prices by less than 1 percent while resulting in $280 billion in savings on consumer natural gas bills between 2013 and 2050.
EPA’s national renewables targets are modest. Lower than EIA’s business as usual (BAU) projection for 2020 and only marginally stronger than BAU in 2030. Image: UCS
NERC’s pessimistic projections for wind, solar and other renewables has a lot to do with consultant EVA’s use of pessimistic cost and performance assumptions for these technologies. The modeling assumes capacity factors for new wind projects will range from 32-39 percent, and are fixed or declining over time. In contrast, recent wind projects using new technology that are being developed in high-quality wind sites are achieving capacity factors of between 40 and 50 percent, and higher.
DOE’s Wind Vision study projects that capacity factors will continue to rise and capital costs will continue to fall over time, making wind even more cost-effective in the future. NERC’s projected costs for solar PV systems are also pessimistic compared to DOE’s study.
Just to show it’s not just renewables that NERC underestimates, the 2025 reference case results for New York-New England shows minimal additions of 115 MW of gas combined cycle generation and in the Mid-Atlantic region, only 808 MW.
So, when NERC yells “Fire…”
NERC says that power plant retirements will follow economics. This is more true in areas with extensive industry de-regulation, and less so in other areas that have regulated rates of return, long-term contracts, and reliability obligations for power plants.
NERC says retirements will harm reliability, and thus a delay of the Clean Power Plan rules makes sense. Because NERC’s numbers reveal the retirements are already happening (and never mentions the vast majority of plants retiring are more than 50 years old), the replacements are already coming. Resistance to modernization is not a strong, long-term strategy to ensure reliability. The Regulatory Assistance Project has a fresh, serious discussion of the reliability continuum.
NERC says its modeling results are “representative of potential outcomes, but may not be what will necessarily happen.” It is unclear whether the NERC-EVA models seek to build capacity first, and then look at the energy and CO2 reductions that result, thus optimize for lowest capacity, or look first for lowest cost CO2 reductions first. The language about modeling capacity, and limits on renewable energy suggest it is not energy or emissions compliance that is optimized in the results.
One more set of numbers
The total energy produced by coal plants changes by 8.2 percent from 2016 to 2030 in one of the NERC models. But when NERC projects a change of 8.2 percent in total use of coal plants going forward and pre-determines the contributions that can be made by renewable energy, there are credibility questions.
In the end, they have used a variety of assumptions and statements to make us think that there’s great trouble ahead for the country under the proposed EPA rule. Now maybe I understand why they didn’t release their numbers with the report. Their case is not helped by looking at the numbers.
NERC has generated more controversy than advice in the Clean Power Plan debate.