Before Relying on a Model, Look at the Assumptions

Last week Governor Abigail Spanberger (D) released a modeling-based energy plan for Virginia claiming it could end the use of natural gas, despite accelerating electricity demand. A year ago, Dominion Energy Virginia tested the same proposition in an integrated resource plan of its own.

Lay the two “no gas” future energy generation outlines side by side, and they seem to be describing different states. Governor Spanberger’sresource mix is a sales pitch for massive new solar and wind investments, backed up by batteries. Dominion reported that if Virginia insists on retiring all the natural gas plants in 20 years, as current law requires, nuclear power should replace them.

Spanberger’s approach embraces the Virginia Clean Economy Act (VCEA) no-gas goal, and Dominion last year was warning against it. They used the same software model, proving once again that all model outcomes depend on the assumptions entered at the beginning. It has been 50 years now since a British statistician famously warned that all models are wrong, even if some are useful.

A key assumption behind Governor Spanberger’s plan is that there is a “social cost of carbon” of $262 for every ton of carbon dioxide emitted by a generator, with no citation for where that projection came from. The Dominion 2025 plan doesn’t mention any such cost. The Virginia State Corporation Commission has never accepted a social cost of carbon amount as a legitimate cost in deciding an application.

The Spanberger Administration energy plan will be cited for years now as gospel truth and used to justify another major legislative push to undercut the independence of the State Corporation Commission. The state’s media will parrot the claim that using natural gas is too expensive without ever admitting that the unproven social cost of carbon is the giant thumb on that scale, along with claims of more disease among Virginians.

Governor Spanberger’s gas-free proposal calls for up to 39 gigawatts of additional utility-scale solar generation, plus more solar from private facilities feeding the grid in a distributed generation contract. Dominion’s 2025 gas-free plan, filed with the SCC, proposed less than half as much, both utility-owned and through distributed generation.

Governor Spanberger’s proposal calls for up to 9.6 additional gigawatts of offshore wind, about 600 more of the type of turbine now being built off Virginia Beach. That would cover hundreds of square miles of ocean, just like her solar wishes would cover hundreds of square miles of land. Dominion also called for more offshore wind, but not that much.

The big difference between the two proposals is that Dominion relied upon adding more than 12 gigawatts of new nuclear generation to replace the 12.7 gigawatts of hydrocarbon generation it must scrap to comply with the Virginia Clean Economy Act’s zero-carbon mandate.

Governor Spanberger and her staff proposed only 2.2 gigawatts of new nuclear in their “no gas” ideal plan. They would replace most of the disappearing natural gas with 12 gigawatts of distributed generation from privately built solar and batteries scattered in thousands of homes, offices, and factories. Pixie dust, in other words.

Some other key points about the Spanberger plan:

• It assumes that the growth of demand from data centers continues, despite all the political controversy about them and all the calls for moratoriums.

• It assumes Virginia will continue to import massive amounts of energy from other states within the PJM Interconnection region, but dances around the inconvenient fact that much of that power comes from carbon-emitting coal and natural gas.

• The first part of the plan document that describes the electricity generation system in the state is full of useful information. Virginia is in for a world of hurt if new generation does not appear soon.

• Buried in the pages are some numbers that can be used to compare the cost and efficiency of the various methods of generation. A prediction published here earlier about the $60 billion-plus costs of the Clean Economy Act’s battery mandate is confirmed. Any future wave of offshore wind costs will swamp the cost of Phase I.

• Whatever gets built, whatever scenario ultimately clears the SCC, the cost of electricity will rise and then rise some more. The Regional Greenhouse Gas Initiative carbon allowances, for example, are predicted to be five times higher by 2040. Detailed bill predictions like what Dominion released last year are absent from Governor Spanberger’s document.

• Two of the scenarios Governor Spanberger blessed in her news conference do call for adding small amounts of new natural gas before the 2045 deadline. Frankly, that just adds to the cost increases because the intention remains to close them long before their useful lives are over. If built, they should stay open past 2045.

For comparison, one of the Governor’s modeled scenarios did assume that the Virginia Clean Economy act and its constraints on natural gas are repealed, an alternative she rejected in her news conference. The Dominion 2025 document did the same, again with a very different result. The utility’s “no VCEA” model proposed only about one-third as much new natural gas as Spanberger’s modelers did.

By tripling the amount of new natural gas, the Spanberger Administration “no VCEA” model also tripled those social cost of carbon and alleged health impacts. It was one more easy-to-flag effort to steer the results in favor of the solar-wind-battery approach.

Governor Spanberger’s effort to prove that the VCEA can be achieved without ruinous cost is very similar to the data the SCC was presented with by opponents of the new Chesterfield County gas generation plant. The SCC in last year’s decision agreed with Dominion that the plant was needed to maintain reliability.

If left alone to do its job, absent changes in the law, the SCC should be able to parse all the conflicting data and ensure reliability while choosing more affordable pathways. Whether submitted by a utility or written by politicians, these energy resource plans are only words on paper until the SCC gets actual proposals.

Spanberger and Dominion Asked the Same Software the Same Question and Got Different Answers

Key Takeaways:

• Assumptions drive the results in any model. The article argues that Spanberger’s $262-per-ton social cost of carbon steers her model toward wind, solar, and batteries.
• The modeling of two gas-free plans offers sharply different paths. Dominion’s model relies heavily on new nuclear power, while Spanberger’s emphasizes renewable generation and batteries.
• Costs and reliability demand independent scrutiny. The article warns of rising electricity costs and argues that the SCC must remain free to evaluate actual projects.

Key Quote:

“If left alone to do its job, absent changes in the law, the SCC should be able to parse all the conflicting data and ensure reliability while choosing more affordable pathways.”

Steve Haner
Jefferson Forum

Steve Haner is the Senior Fellow for Energy and the Environment at the Jefferson Forum and may be reached at Steve@thomasjeffersoninst.org