Nuclear subsidy briefs, August 15, 2019
Nuclear economics remain underwhelming, but the creative ways the industry pursues to shift risks onto others and to seek new subsidies for themselves continue unabated. Here's a run-down of a few recent developments.
1) NRC approves sale of closed Pilgrim nuclear reactor to Holtec. After studying the deal, NRC decided Holtec can do this, even though they've never done it before and are trying to do it for the first time in lots of places at once. Absent adverse rulings by commissioners in the next couple of days, the decision will become final. Or mostly, sort of, final. Holtec gets the closed contaminated reactor, the land, and more than $1 billion in the plant's nuclear decommissioning trust fund. That money will be used to pay for decommissioning -- much of it conducted using staff, facilities, and resources related to the plant's new owner or affiliated parties. The NRC decision also gives Holtec an exemption to use the decommissioning trust fund to pay to manage the spent fuel and other site restoration activities. What could possibly go wrong?
Not surprisingly, Holtec's spokesperson Joe Delmarin concluded that "the transfer of Pilgrim to Holtec for prompt decommissioning is in the best interests of the town of Plymouth and surrounding communities, the nearly 270 people from the region who work at Pilgrim, and the Commonwealth." A bit early to tell, in my view. And the Commonwealth doesn't seem to agree either.
Detailed and well-structured petitions to intervene by both citizen's group Pilgrim Watch and the Massachusetts Attorney General seem largely to have been ignored. Though, here the "mostly, sort of" framing comes into play, as the NRC writes in its decision that:
These requests are pending before the Commission. The hearing, if granted, will not be completed prior to approval of the license transfer application. The order approving the transfer will include a condition that the NRC staff’s approval of the license transfer is subject to the Commission’s authority to rescind, modify or condition the approved transfer based on the outcome of any post-effectiveness hearing on the license transfer application.
Mary Lampert, director of Pilgrim Watch, puts little stake in these caveats, however. And there is no indication in the current license transfer approval that the NRC is ensuring backup funding from the current owner should Holtec run out of money. Since the new owner has structured the deal using poorly capitalized stand-alone limited liability companies, such prudence would be warranted.
For more on Holtec and the potential problems on their closed reactor buying spree, see a deep-dive on the issue I did in an earlier blog post.
2) Faster decommissioning means moving lots of radioactive waste somewhere; and those places don't seem happy to get it. The Permian Basin is undergoing and oil and gas boom. It is also being targeted for "interim" storage of a variety of nuclear wastes. Not everybody is happy about this. The Texas facility would be run by Interim Storage Partners, a partnership of Waste Control Specialists and Orano USA. Orano, formerly Areva Nuclear Materials, is involved with the rapid decommissioning of Vermont Yankee -- which is not a Holtec project but using a similar approach.
Holtec's go-to holding ground for nuclear waste from its projects would be in southeastern New Mexico. Holtec's facility is also tagged as "interim," though few seem to believe that. "Interim," after all, is a fairly flexible term: with a radioactivity window of thousands of years, one can call a facility interim with a straight face, knowing full well stuff will be sitting there for centuries. Opponents to Holtec's waste facility are many, and include New Mexico's governor.
How rejection, long delays, or much higher costs in the disposal sites affect the economics of Holtec's ability to carry out the decommissioning at the reactors it now owns without running out of money will be interesting to see.
3) Shellenberger: summer is for nukes. Most people think beaches and gin and tonics during summer. For Michael Shellenberger, winter, spring, summer or fall doesn't matter: nuclear power is always on his mind. He is a dogged advocate of a nuclear-led solution to climate change. And while I often admire his dedication, he sometimes looks only on the bright side of life, glossing over important and material limitations with his favored nuclear solutions in the process. This seems to have been the case in his promotion of nuclear reliability in summer as a big selling point of the technology in a column he wrote last summer in Forbes, and that I recently re-read.
He rightly pointed out that peak electricity demand is often during summer, and shortages are common. His specific examples focus on plant outages, a lack of wind during periods of high heat, and solar panels' reduced efficiency at very high temperatures as factors that can trigger regional power supply deficits, and surging peak prices. He argues that particularly in summer, nuclear is central to rescuing the grid, pointing to South Korea, Taiwan, and Japan as examples where reactors, including restarting closed ones, were integral to meet summer demand in recent years.
And in his own state of California, Shellenberger references the National Weather Service telling people to go somewhere else if they lose power. He also argues that power is much more expensive once reactors close. This last point raises once again a core question I've been trying to get a good answer to for years: if nuclear really (and not just in industry PR packs) helps keep energy costs so much lower for all users, why is it that the industry can't figure out a business model that keeps operating reactors solvent without crying for bailouts in state after state?
But that is a question for another day. Back to Shellenberger's nuclear summer: in pointing out a handful of countries relying on nuclear to meet summer peaks, it is hard to believe that somebody as prolific in this area as he is would be unaware of the quite significant counter-case to his core argument. Far from always being the go-to resource to beat the heat, reactors in many parts of the world have been shuttered during summer months because they need massive amounts of cooling water. And the problem appears to be getting worse.
High temperatures can reduce available intake flows and increase the ambient temperatures in receiving waters. The high temps already stress wildlife even before the water is further heated by doing reactor-cooling duty. At a certain point, the reactors are no longer able or allowed to pull water or to discharge extra heat. Depending on the capacity of the receiving waters, reactors may need to be fully shut down; in other cases they must ramp down to a lower production level.
Ironically, one of usually-ignored subsidies to reactors (at least in the US, though likely in other countries as well) is free cooling water. Not only free, but also generally senior and firm water rights, sometimes at the expense of other users of that same water body. In addition to reducing the delivered cost of nuclear electricity, this subsidy also allowed reactor owners to largely ignore cooling efficiency when they designed and built the current generation of reactors. Had the reactors been forced to pay for their water rights, they likely would have adopted much more water-efficient cooling techniques, and as a result been less likely to face forced closures during periods of high heat now.
Here are a few links that highlight a much more complicated picture on summer nukes than the one Shellenberger painted. The first is an overview of the issue and how it will get worse with rising global temperatures by Christina Chen at NRDC. An earlier paper by Linnerud, Mideska and Eskelund looks at how higher temperatures in ambient cooling water can reduce reactor efficiency. Plant closures due to cooling water issues include France, Sweden and Finland (2018); and France and Germany (2019). Heat-related curtailments, discharge heat waivers, and shutdowns have been common in earlier years, including in the United States (2003, 2012, and probably other years as well were I to keep searching). US nuclear power curtailments have been rising over time, along with global average temperatures. Interestingly, I couldn't find examples on nuclear power curtailment during hot weather outside of the US and Western Europe. This either means that the reactors in all of the other countries have much better proximity to high-cooling capacity receiving waters; that they have weaker regulations in terms of heat discharge; or that there are curtailments and closures, but they are not being reported.
4) NuSubsidies I - fast breeder reactors. Federally funded. High costs, increased proliferation risks. I'm sure it will go better than last time.
5) NuSubsidies II - advanced reactors. Forty-year power purchase agreements allowed (max of ten years for other technologies); can be at above market rates in the Nuclear Energy Leadership Act (NELA), reintroduced in May. Apparently targeted at small modular reactors, though the bill language appears flexible enough to apply to derivative generations of current, larger reactors as well so long as they meet the NRC licensing window set out in the proposed legislation. Thankfully, this is still proposal, not actual law.
6) And the band plays on: more delays and cost escalation for the Vogtle reactors. Not surprising at this point to see the Vogtle monster continue to burn cash, though the never-ending cost increases are a bitter pill for ratepayers on the hook to buy the power at the end. If the project goes bankrupt (as it still could), the big losers shift from ratepayers to federal taxpayers. As of March 2019, the Department of Energy boosted the size of taxpayer loan guarantees to $12 billion. Some day, Congress or the DOE Office of Inspector General will finally do a real investigation on how this fiasco emerged. That investigation may well include a look into the use of private emails for public business.
Even before financing costs, the two-reactor set is expected to exceed $17 billion in cost. More delays are also expected, and when one has a very large investment for which revenues keep getting pushed further into the future, the financing costs will continue to escalate quickly. The end of the complex portions of the construction effort are not near either:
Already, the remaining work is expected to be 2.5 times more dollar-intensive than the construction finished thus far. Workers have completed 77 percent of the project at a cost of $9.86 billion. The remaining 23 percent is budgeted to cost $7.243 billion, assuming it does not run over.
This summary, from two weeks ago in a Greentech Media article by Julian Spector, ends with a quote from Jessica Lovering, director of energy at The Breakthrough Institute, a group that has been pushing nuclear as a central (often the central) solution to climate since its founding more than 15 years ago. Lovering remarks "Ultimately, we need to move away from nuclear power plants as these big infrastructure projects with lots of people doing artisanal craftsmanship and move toward factory fabrication."
I've never heard the term "artisanal craftsmanship" applied to massive nuclear projects before. But I kind of like it. It makes a huge, expensive, and too often poorly-run engineering undertaking sound like a quaint craft fair.
Yes, reactors, or large one-off construction projects in general, are prone to a certain set of problems that drive costs up. But it is also true that for 60 years, cost reductions in nuclear power came through economies of scale to spread massive costs over a massive number of kWh. Indeed, these assumptions on where economic gains were to come from continued to play a central role in some of the core research by MIT and the University of Chicago that underpinned the planned nuclear renaissance in the early 2000s. These analyses suggested that while your First-of-a-Kind (FOAK) unit might be expensive (though their projections were way lower even on FOAK than what has emerged), the "nth of a kind" reactor would be cheaper than competing forms of baseload power.
Standardization and production lines, Lovering's new model, really seem to apply only to the small modular units. The standardization could lead to lower unit costs of production; but the scale of savings seem far too small to really bring LCOE in line with alternatives. Indeed, high cost seems likely to remain a problem even with the SMRs for quite some time to come.
How long the period of high costs will last remains a critical variable: in the end, SMRs are competing less against wind or solar, and more against the power storage that converts intermittent resources into generation just as reliable as nuclear for the vast majority of applications. Could SMRs ramp up their learning curves on both technology and production quickly enough to be relevant in dealing with climate change over the next 10-15 years? Bill Gates certainly believes so. But I view it as a long-shot, as the costs of power storage continues to fall more rapidly than expected and the market demand for batteries remains enormous, driving massive research and experimentation on all elements of the storage product cycle. Four years after I wrote this this, power costs are falling faster than predicted, and nuclear costs continue to rise. My bet is still on batteries to win.