Why does Duke Energy’s carbon plan shortchange solar?

For months, Duke Energy has been defending its carbon-cutting plans for its core Carolinas service territory against critics who say the utility’s energy forecasts deliberately understate how much solar power and battery storage will be available to the grid. Now, the utility is making its case in person to an audience of regulators and skeptical clean-energy advocates.

Last week, the North Carolina Utilities Commission began what is expected to be weeks of expert testimony and cross-examination on Duke Energy’s plan to comply with HB 951, the North Carolina law that requires subsidiaries Duke Energy Carolinas and Duke Energy Progress to cut carbon emissions by 70 percent below 2005 levels by 2030.

Duke is asking the commission to consider four different proposals — only one of which actually aims to meet the 2030 carbon-reduction target rather than pushing it out to 2032 or 2034, and all four of which call for building about 2 gigawatts of new fossil-gas-fired power plants by the end of the decade. Environmental activists, clean energy trade groups, consumer advocates, local governments and the state’s Attorney General’s Office have all rejected Duke’s proposals and submitted their own alternatives to the commission, which they say will not only reduce carbon emissions more quickly but also save money over the long run.

The reason for this, according to Tyler Norris, vice president of development at solar developer Cypress Creek Renewables, is simply that ​“solar and solar-plus-storage are the most affordable, proven and scalable zero-carbon resources available to Duke’s system.”

Duke argues that its plans are the safest and most cost-effective option and that the alternative plans are ​“based on flawed and incomplete analysis.” The groups challenging Duke have pushed back, saying that the utility’s lack of transparency about the data and methods that informed its conclusions has left them no option but to create their own plans.

Utilities and state regulators across the country face similar challenges as they strive to cut emissions at a pace needed to stave off the worst impacts of climate change. Half of the states in the country have set greenhouse-gas reduction targets, and many utilities in states that haven’t set targets have set their own voluntary goals, as Duke did in 2019 with its commitment to hit net zero by 2050.

But Duke, which owns one of the country’s largest and most carbon-intensive generation fleets across multiple states, could play an outsize role in cutting electricity-sector emissions, a key step in decarbonizing transportation, heating and other sectors.

The North Carolina Utilities Commission is required by statute to choose a plan by December 31. To make that decision, its seven members must weigh a staggeringly complex set of variables, from the future costs and availability of emerging technologies such as small modular nuclear reactors and green hydrogen to how much support Duke should provide for energy efficiency, demand response and rooftop solar.

But amidst a sprawling and long-running debate over North Carolina’s clean energy future, the biggest point of contention in the near term is whether Duke Energy is intentionally downplaying the role that utility-scale solar and batteries can play in its transition to a low-carbon grid.

Caps on adding new solar: A major sticking point 

Duke’s critics believe that the utility is doing just that. One of the many complaints they’ve raised involves the utility’s approach to modeling a cost-effective future grid mix — in particular, its decision to insert a cap on how much new solar its computer models could select between 2026 and 2030: a limit of 750 megawatts in 2026, 1,050 megawatts in 2027 and 1,350 megawatts per year from 2028 onward.

That’s a critical limitation on modeling that should be used to inform how the utility and regulators can best balance the priorities of reducing carbon emissions, maintaining reliable grid service and minimizing the rate increases customers will bear to pay for it all.

A study by Synapse Energy Economics commissioned by groups including the North Carolina Sustainable Energy Association, the Southern Alliance for Clean Energy, the Natural Resources Defense Council and the Sierra Club found that a more solar-heavy portfolio could save $700 million by 2030 compared to Duke’s plan while maintaining reliable electricity supplies. Expanding access to wind power from the Midwest — something outside the scope of Duke’s plans — could bring much greater savings, the analysis found.

“We’re advocating for a plan that successfully achieves the 2030 mandate of 70 percent carbon reduction in a way that has minimal impact for ratepayers,” said Matt Abele, the North Carolina Sustainable Energy Association’s marketing and communications director.

A study by The Brattle Group commissioned by the Clean Power Suppliers Association trade group found similar results. Allowing models to select solar beyond Duke’s caps adds 7,500 megawatts of solar by 2030, compared to Duke’s 5,175 megawatts. This reduces the total annual system costs of the portfolio by $860 million in 2030, $730 million in 2032 and $690 million in 2035, for cumulative savings amounting to billions of dollars.

Duke will have opportunities to refine its longer-term plans with the commission in the years ahead, Kendal Bowman, Duke Energy Carolinas’ vice president of regulatory affairs and policy, said during the first day of hearings on September 13. At the same time, ​“we’re asking for a suite of near-term procurement decisions and actions” to set the stage for building what’s needed in the years between now and 2030, she added.

Capping the amount of solar and solar-plus-battery systems that can be added between now and 2030 affects those near-term plans, most notably by leaving a hole in Duke’s generation portfolio that must be filled as Duke closes down its Carolinas coal plants by 2030 — a hole that Duke plans to fill with fossil gas.

That’s why ​“there’s arguably no single modeling assumption in dispute in this proceeding with greater implications than the solar interconnection constraint,” Cypress Creek Renewables’ Norris said in an email. ​“Duke’s proposed solar cap [will] make it extremely difficult, if not impossible, to achieve North Carolina’s CO₂ reduction goal by 2030, while significantly increasing compliance costs and forcing reliance on more expensive technologies.”

Conflicts over modeling 

Duke claims that moving more quickly to cut carbon will raise costs for customers. According to its carbon plan, Portfolio 1 — the only scenario that achieves a 70 percent reduction in carbon emissions by 2030 — will cost $2 billion more than its Portfolio 2 option, which delays hitting the 70 percent target until 2032, and $6 billion more than Portfolios 3 and 4, which don’t reach the target until 2034.

Electricity prices are important to all customers, and we’re doing all that we can to keep these rates affordable,” Bowman said in last week’s testimony. 

But the North Carolina Utilities Commission has already ordered Duke to revise some of its modeling after critics pointed out flaws earlier this summer. Tyler Fitch, a senior associate with Synapse Energy Economics, noted in his testimony to the commission that Duke manually overrode its initial modeling in ways that replaced about 1 gigawatt of battery storage capacity with gas-fired combustion turbines in its 2035 models, and replaced an even greater amount of batteries with gas and nuclear power by 2050 (see chart below).

To comply with the commission’s request, Duke filed two supplemental portfolios in late July. Those portfolios include changes welcomed by some critics, such as revising how batteries and solar-battery systems are modeled so they charge with lower-cost energy and discharge when energy prices are higher, increasing their economic value.

But the two supplemental portfolios don’t lift the caps on solar deployments from 2026 to 2030. Earlier this month, North Carolina Attorney General Josh Stein weighed in, asking the commission to reject Duke’s plan and to decline any proposals from Duke to streamline permitting new gas plants before ​“no-regrets” solar, battery and onshore wind deployments can be pursued.

The Attorney General’s Office hired consulting firm Strategen to develop an alternative plan that Stein called a ​“stronger, smarter and more affordable” option. Edward Burgess, a senior director at Strategen, testified to the commission that it should be wary of accepting Duke’s supplemental portfolios as a ​“middle ground” between Duke’s original proposal and alternative plans since the new portfolios don’t attempt to reach the goal of 70 percent emissions reduction by 2030 and ​“fail to address concerns raised by other parties” — including lifting its caps on new solar.

Interconnection is a problem. Are there solutions? 

Since filing its initial carbon plan in May, Duke has maintained that its cap on adding new solar and solar-battery systems is based on an undeniable real-world constraint: the difficulty of adding so much new generation capacity to a constrained power grid.

Members of Duke’s planning and analytics team reiterated this point during last week’s opening day of testimony. Duke’s caps ​“reflect what real-world realities are,” said Glen Snider, Duke’s managing director of planning and analytics. ​“We have constraints on every single resource, and not just on solar.”

Interconnection constraints have become a major bottleneck for renewable energy across the country. Still, critics say that Duke has failed to back up its assertions with data and instead has cited ​“internal engineering judgments” as the primary justification for imposing them.

During last week’s hearing, Benjamin Snowden, an attorney representing the Clean Power Suppliers Association, noted that Duke’s proposed cap of 750 megawatts in 2026 and 1,050 megawatts in 2027 are roughly equivalent to the amount of solar that Duke reports interconnecting in 2015 and 2017, respectively.

This implies that the utility expects to have made no improvements to its interconnection process over a decade’s time, despite the potential for those improvements to create ​“opportunities for ratepayer savings,” he said.

One big question for Duke is how quickly it will be able to expand its transmission grid to make room for new solar. The utility plans to invest $560 million in grid upgrades in what it has identified as its ​“red zone,” a swath of North and South Carolina where grid constraints have delayed nearly two-thirds of the approximately 6,000 megawatts of solar already awaiting interconnection.

Norris said in testimony that this grid expansion, along with improvements in Duke’s interconnection process, should allow the utility to ​“achieve significantly greater improvement over its historic solar interconnection rates.”

But upgrading transmission grids is a notoriously difficult and time-consuming task, which could restrict how quickly Duke can execute its red-zone expansion plans.

Is Duke’s carbon plan designed to protect its revenue stream? 

So why have Duke’s plans and modeling processes consistently undervalued solar and solar-plus-battery installations in favor of other generation technologies? It may be because solar and solar-battery systems are the only type of technology that Duke Energy isn’t legally allowed to build completely on its own, add the costs to its ​“rate base” and then recover those costs by hiking the rates of its customers, multiple critics say.

North Carolina law requires Duke to open up 45 percent of solar and solar-battery projects to competitive bids from third-party developers. ​“In other words,” Norris said in his testimony to the North Carolina commission, ​“for every megawatt of solar or solar-plus- storage that is selected, Duke is forgoing an opportunity to own and rate-base additional capacity.”

That ​“cost-of-service” model also incentivizes Duke to prefer more expensive generation technologies over less-costly ones, Norris said. Duke’s longer-range plans call for significant investments in offshore wind, a resource that has yet to be built off the Carolinas coastline, and small modular nuclear reactors, a class of nuclear power that has yet to be built at commercial scale. Both of these have much higher costs per megawatt of generation capacity than solar or solar-battery systems.

“Duke maintains that the best approach is to make conservative assumptions about how much solar it can interconnect,” Norris said in testimony. But the result would ​“massively increase earning opportunity for Duke’s shareholders at the expense of ratepayers by driving selection of higher-cost resources.”

Duke has harshly criticized the cost forecasts its critics have presented in their alternative plans. In testimony to the commission, members of the utility’s planning and analytics team noted that those cost forecasts have higher projected costs for combined-cycle and combustion-turbine gas power plants, and lower costs for solar, batteries and onshore wind turbines, than those used by Duke.

These and other discrepancies indicate that alternative plans ​“unduly favor certain resources” and ​“lack prudent diversification of risks,” utility representatives testified.

But Duke critics say that the consistency of Duke’s approach to modeling its future energy needs in ways that would allow it to build more of the assets that earn it more money — including those approaches that have been found to be flawed and ordered to be redone — reveal an intent to skew the results to help secure future financial gain.

“This goes well beyond honest differences of opinion,” Jim Warren, executive director of advocacy group NC Warn, said in a July statement. ​“All Duke’s ​‘errors’ coincidentally fall in its favor.”

Duke officials deny that the utility purposefully manipulated its plans for its own financial benefit. On the first day of last week’s hearings, John Burns, general counsel of the Carolinas Clean Energy Business Association, asked Glen Snider, Duke’s managing director of planning and analytics, pointedly – if indirectly — whether that was the case:

“Is it your intention, Mr. Snider, that Duke Energy’s work on the carbon plan is not outcome-based in any respect?” 

“We took inputs with no desire to say a particular resource is going to be favored,” Snider replied. 

Forecasting the future cost of different energy resources is an inherently uncertain task. But multiple critics of Duke’s plan noted that last month’s passage of the Inflation Reduction Act, which extends and expands federal tax credits for solar, wind, stand-alone energy storage and other carbon-free resources, will require the utility to recalculate — and likely revise downward — how much it believes those resources will add to customers’ future cost burdens.

As the Attorney General’s Office pointed out in a filing with the commission, the cost reductions from the new law mean that ​“the proposals now under consideration in the Carbon Plan proceeding are already out of date.”

It’s not clear, however, if regulators are prepared to demand that Duke rerun its models with new cost forecasts or any other changes, given that they have less than four months to meet the statutory deadline to choose a carbon plan.

In August, the commission rejected a request made by the Attorney General’s Office to require the utility to rerun its modeling to correct what the office described as ​“important and substantial modeling flaws.” The commission’s order said that ​“would not be reasonable at this point in time given the compressed timeline for this proceeding” and placed the burden of proving the models were flawed on those seeking to ​“rebut Duke’s proposed portfolios.”

In the next few weeks of hearings, it will be crucial for clean-energy advocates to make their case that the utility’s plan is flawed and that their alternatives are a better choice for the commission, North Carolina Sustainable Energy Association’s Matt Abele said. So far, Duke has been ​“continually digging in…on their plan as proposed,” he said. ​“It’s up in the air as to what outcomes or results that cross-examination might yield.”