By Rich Heidorn Jr.

WASHINGTON — The 12,000 attendees at the World Gas Conference last month seemed unaware that the ground is shifting under their feet. Yet that is what’s happening, the Rocky Mountain Institute contends in a new analysis.

The clean energy think tank says that utilities, investors and regulators should be skeptical of any new investments in natural gas-fired generation because the combination of renewables and storage is already cheaper than combustion turbine peakers in some regions and will fall below the cost of combined cycle plants within a decade.

RMI’s analysis is the latest to sound warnings for gas. In March, IHS Markit published an analysis that found that batteries with access to cheap renewable power can be cheaper than CTs.

Greentech Media Research says storage will be competitive with gas peakers within four years and cheaper within 10 years. “I can’t see a reason why we should ever build a gas peaker again in the U.S. after, say, 2025,” Shayle Kann, a senior adviser to GTM Research and Wood Mackenzie, told GTM’s Energy Storage Summit last December. “If you think about how energy storage starts to take over the world, peaking is kind of your first big market.”

However, Bloomberg New Energy Finance does not see gas’ role as the “bridge” fuel between coal and renewables ending any time soon, instead forecasting an increased role for gas peakers for the next three decades.

At the World Gas Conference at the Walter E. Washington Convention Center, speakers on a panel on the role for gas in power generation also were more bullish.

With global power consumption expected to double by 2050, gas has a “great opportunity” to grow, said Shankari Srinivasan, vice president and managing director of global gas and EMEA power for IHS.

She acknowledged the growing competition from renewables and batteries and the impact energy efficiency can have on power demand growth. But in constructing future scenarios, IHS “found it very difficult to construct a case … where [global] gas-fired generation declines,” she said. “Renewables on their own are unlikely to be able to support this level of growth. In the U.S., I think we will continue to see gas taking a large share of power generation and remaining a fundamental part of the power generation mix.”

In contrast, Srinivasan said Europe may be seeing “the beginning of the end” of combined cycle gas turbine construction.

“I am, I think, being a little provocative. But it is possible to envisage new generation capacity as a mix of renewables and maybe only open cycle gas turbines,” she said.

The price of gas will be paramount in China and the rest of Asia, which will each account for one-third of global power demand growth through 2050.

“In Asia, development and growth of gas will depend on how competitive it is with coal … and the strength of clean air policies. … Gas as a bridge to a zero-carbon future may be skipped entirely — replaced by a gray-green world of coal and renewables in certain parts of the world.”

Another panelist, De la Rey Venter, Royal Dutch Shell’s executive vice president for integrated gas ventures, insisted that gas will remain essential to balancing the variability of renewables for at least a decade.

“There are those who say that ultimately batteries will eat gas for breakfast. We don’t quite subscribe to that logic,” he said. “There are many things that need to happen before batteries can play a meaningful role beyond short-term balancing of fluctuations in the system.”

In addition to reducing their cost and improving their performance, Venter said batteries need frequent access to low- or no-cost power for charging to be competitive.

Venter said the gas industry needs to “fight this notion that …. there’s this existential competition between gas and renewables.”

“Gas is the ultimate enabler of renewables,” he said. “If you really want to see a … widespread penetration of renewables, you need to have gas for the next decade or two.”

Stakes: Replacing Half of Thermal Capacity by 2030

Although there is disagreement over when gas will lose its appeal for power generation, there’s no doubt that the stakes are huge, both for the industry and consumers.

RMI notes that more than half of U.S. thermal generating capacity is more than 30 years old and expected to reach retirement age by 2030. It estimates that it would cost more than $500 billion to replace all retiring power plants with new natural gas-fired capacity (including $110 billion in investments already announced by utilities and independent power plant developers).

“This will lock in another $480 billion in fuel costs and 5 billion tons of CO₂ emissions through 2030, and up to 16 billion tons through 2050,” RMI says. “The current rush to gas in the U.S. electricity system could lock in $1 trillion of costs through 2030.”

RMI sees a $350 billion (net present value) market opportunity through 2030 for renewables and distributed energy resources supplanting gas projects where cost effective. That would eliminate $370 billion of gas capital costs and operating expenses, a net savings of more than 2%, it said.

“This investment trajectory would unlock a market for renewables and DERs many times larger than today’s,” RMI said. It would also reduce carbon emissions and save consumers money — even excluding DERs’ value to the distribution system beyond peak load reduction or avoided fuel price risk or any emission costs.

Cheaper than Peakers, Nearing Parity with Combined Cycle

RMI’s analysis found that the clean energy portfolio — wind, solar and DERs, including batteries — was cheaper than two CT plants planned for serving peaks, beating one in the Mid-Atlantic by 60% and one in ERCOT by 47%.

In a comparison with CCGT power plants with higher capacity factors, RMI said the clean portfolio was 8% cheaper than a CCGT in California but 6% more costly than such a project in Florida, RMI said.

“Factoring in expected further cost reductions in distributed solar and/or a $7.50/ton price on CO₂ emissions, all four cases show that an optimized clean energy portfolio is more cost-effective and lower in risk than the proposed gas plant,” the report said.

In addition to competing with proposed gas generation, clean energy portfolios will also undermine the profitability of existing plants within eight years, RMI says. In some areas, clean energy portfolios’ combined construction and operating costs — levelized cost of electricity — will be lower than CCGTs’ operating costs by 2026, assuming $5/MMBtu gas (translating to operating cost of $36/MWh). Assuming gas prices remain about $3/MMBtu ($23/MWh), the alternatives won’t be cheaper until about 2040 — still within the operating lives of plants being proposed now, RMI says.

$144 Billion Stranded?

“In other words, the same technological innovations and price declines in renewable energy that have already contributed to early coal plant retirement are now threatening to strand investments in natural gas,” the report says. “Thus, the $112 billion of gas-fired power plants currently proposed or under construction, along with $32 billion of proposed gas pipelines to serve these power plants, are already at risk of becoming stranded assets.”

With about 83% of announced gas projects proposed for restructured markets, independent power producers would bear most of the risk of competition from DERs and renewables.

The trends are also beginning to pinch IPPs’ suppliers. Bloomberg reported in June that Siemens is considering selling its gas turbine business. The company’s CFO told investors in March that the market for large gas turbines will fall to 100 units in 2018, 10% below the company’s previous projections.

Competitor General Electric also sees a “soft” market for gas turbines for several years. Two-thirds of its power capacity additions in 2017 were renewables. But in announcing a corporate restructuring in June, the company told investors, “Gas remains key to long-term energy mix.”

California, Arizona Leading the Transition

RMI cites as examples 11 alternatives to new thermal power plant investment now under consideration. Six of them are in California, where the abundance of solar and wind — and the state’s environmental goals — have made gas-fired generation an endangered species.

In February, the California Public Utilities Commission issued its first integrated resource plan. Intended to help the state meet its 2030 greenhouse gas reduction goals — a 50% reduction in electric sector GHG emissions from 2015 levels — the plan sees no new gas-fired capacity through 2030. Incremental generation needs are instead satisfied by utility-scale solar (73%), in-state wind (9%), battery storage (16.3%) and geothermal (1.7%).

CAISO has approved battery energy storage (BES) as a capacity resource if it can maintain its rated output for four consecutive hours over three consecutive days.

NRG Energy last October asked the California Energy Commission to suspend its review of the proposed 262-MW Puente plant in Oxnard after commissioners recommended rejecting the application. The turnabout came following criticism that the 2014 analysis that supported the gas addition did not reflect steep price declines since then for non-emitting alternative resources. (See NRG Signals Pull-out on Proposed Puente Plant.)

California regulators in January ordered Pacific Gas and Electric to solicit energy storage, renewables and load management options to replace three uneconomic Calpine gas peakers. On June 29, PG&E proposed to fill its need with four storage projects totaling 567 MW.

In March, PG&E solicited proposals to develop up to 45 MW of “clean energy” resources, including at least 10 MW of energy storage, to replace the aging 165-MW Dynegy Oakland jet fuel-fired power plant. It would be the first time PG&E used clean energy resources as an alternative to fossil fuels for transmission reliability. (See PG&E to Seek Storage, EE to Replace Dynegy Plant.)

Also in March, the Arizona Corporation Commission rejected Arizona Public Service’s plans to double its natural gas fleet over the next 15 years, instead ordering that utilities show that storage is not a cost-effective option before seeking approval of new natural gas units.

APS and Xcel Energy Colorado are among the utilities whose solicitations have produced renewable and storage bids at lower energy or capacity costs than thermal generation. Xcel received 87 bids for solar/storage projects at a median price of $36/MWh — compared with the $85/MWh levelized cost of electricity for an advanced CT, according to the Energy Information Administration.

To be sure, battery technologies will have to improve to be a solution in colder climates, where winter peaks can last for more than four hours.

Nevertheless, Xcel Energy CEO Ben Fowke told The Wall Street Journal earlier this year, “I could see in 10 to 15 years where you have 30% of what is traditionally a peaker market served by storage.”

Mark Dyson, one of the authors of the RMI study, discussed its findings in a webinar last week, citing evidence that investors agree with its conclusions.

Dyson cited the 7% one-day drop in GE’s share price in late May after CEO John Flannery told investors that the market for the company’s large gas turbines will remain weak through 2020. “The narrative was around the bad bet that it made in doubling down on new gas as a growth opportunity,” Dyson said.

“We see other investors looking at the PJM capacity market results and seeing how much uncleared gas there was that was in the queue. That’s kind of another hint that the market is cooling,” he added.

IHS also Bearish on Peakers

The results of RMI’s analysis were consistent with those published in March by IHS Associate Director Wade Shafer and senior analyst Sam Huntington.

The two compared a scenario in which California’s incremental resource adequacy needs from 2021 to 2030 were met entirely with CTs versus one using four-hour lithium-ion (Li-ion) BES.

They concluded that despite projected cost declines, the levelized fixed cost (capital and fixed operations and maintenance costs) of a four-hour Li-ion BES system will remain more expensive than a typical CT through 2030. But with inexpensive power for charging, they said, batteries would have a lower operational cost. “If the savings in systemwide production costs exceed the premium in fixed costs, BES systems would yield net benefits relative to CTs,” they said.

IHS’ analysis assumes that by 2030, more than half of California batteries will be linked to otherwise-curtailed solar PV, giving them access to low- and no-cost power.

If California met all its peaking capacity needs from 2021 to 2030 with BES instead of CTs, net present value benefits to the power sector would be about $16 million — essentially break-even given the size of the investments, the study found. “Savings also arise from the higher efficiency of the remaining thermal fleet — batteries smooth the variability in net load, resulting in fewer start-ups by peakers and allowing mid-merit plants to operate at lower heat rates,” they wrote.

Their conclusion: “California appears to be on the right track in terms of requiring batteries to cost-effectively manage the excess solar energy created by the [renewable portfolio standard]; however, IHS Markit has not evaluated the optimal year to fully transition from new gas-fired capacity to batteries.”

In an interview, Huntington said the RMI analysis appeared sound. He said he was somewhat surprised that RMI found not only peakers but CCGTs at risk. “A lot [of the RMI analysis] relied on energy efficiency and demand response, something we haven’t looked at as closely,” he said.

Dissenting Voices

In contrast, BNEF contends gas will remain vital through 2050.

Its 2018 New Energy Outlook report predicts coal and nuclear will “have almost disappeared from the electricity mix” by 2050, while renewables penetration will reach 55%. Supporting renewables, batteries will “grow in significance” beginning in 2030, it said.

BNEF projects PV module prices to continue dropping at the 28.5% “learning rate” of the last 40 years — meaning costs drop 28.5% for each doubling of deployed capacity.

Li-on battery pack prices will fall by almost two-thirds between 2017 and 2030, BNEF says, driven by the learning rate of a 27-fold increase in electric vehicle sales.

But while cheaper renewables and batteries will hurt most thermal power sources, BNEF sees an increased role for gas peakers.

“As thermal plants retire and variable renewables increase the variability on the supply side … peaking gas emerges as a critical technology to back up renewables during the extremes when wind and solar are at a minimum (sometimes this can be up to weeks at a time),” BNEF said. “We expect peaker gas to grow by almost a factor of four by 2050, as a cheaper, more nimble alternative to large-scale CCGT and coal-fired power plants running at low capacity factors.”