Odd Lots
Odd Lots

Jigar Shah on the Three Big Things Driving the Nuclear Energy Revival

Earlier this month, we got the surprising headline that the shuttered nuclear reactor at Three Mile Island will be restarted. Of course, Three Mile Island was the site of a famous disaster in 1979 — one of the incidents that contributed to the US pulling back on the construction of new nuclear plant

Featured Speakers

Bloomberg HostJigar Shah Guest

Topics Discussed

Episode Summary

Executive Summary: The episode centers on the apparent resurgence of U.S. nuclear power, using Constellation’s deal to restart Three Mile Island as the springboard. Jigar Shah argues that falling policy and financing barriers, rising electricity demand from data centers and electrification, and standardized, repeatable projects could make nuclear economically and financially viable again.

Main Topics: Three Mile Island restart as a nuclear revival symbol (Priority: 5/5): Hosts discuss Constellation’s deal with Microsoft to restart the Three Mile Island site, framing it as a powerful symbolic and market-moving moment for U.S. nuclear power. What changed: policy, demand, and economics (Priority: 5/5): The conversation weighs whether the nuclear comeback is driven more by IRA incentives and federal support or by renewed electricity demand from data centers, manufacturing, and electrification. How IRA tax credits make restarts financeable (Priority: 5/5): Jigar Shah explains that production and investment tax credits, plus bonus credits, can materially improve project economics and make restarts and new builds bankable. Project repetition, workforce, and cost reduction (Priority: 4/5): A major theme is that nuclear costs stay high when projects are one-offs; repeated builds, standardized designs, and retaining trained labor would create learning-curve efficiencies. Grid constraints and interconnection bottlenecks (Priority: 4/5): The episode details how adding generation is slowed by transmission studies, capacity limits, and regional grid rules, which can delay projects for years. New financing and off-take structures (Priority: 4/5): The discussion covers how long-term contracts, contract-for-differences arrangements, and supportive signals from banks and hyperscalers help de-risk nuclear investments. Broader clean-firm energy ecosystem (Priority: 3/5): Nuclear is placed alongside geothermal, hydro, and storage as part of a broader push for clean firm power that can support load growth and industrial expansion.

Key Arguments: Nuclear’s revival is not just about policy; it is also about a real surge in power demand from data centers, manufacturing, and electrification. The IRA’s technology-neutral credits and bonus incentives can make a restarted reactor financially attractive and easier to finance. The biggest reason nuclear got expensive was not only regulation; it was the loss of repeat-build momentum, which destroyed learning-curve cost declines and workforce continuity. Standardization matters: building many similar reactors would lower costs, improve labor productivity, and reduce engineering inefficiencies. Restarting existing sites is easier than greenfield builds because the land, security perimeter, and some infrastructure already exist. Utilities are under pressure to shift from slow-growth dividend models to growth-oriented businesses because of load-growth needs. Long-term off-take agreements from companies like Microsoft convert uncertain market revenues into bankable projects. Grid interconnection delays reflect real engineering constraints, not just bureaucratic sluggishness, since supply and demand must remain in perfect balance across the network. Enhanced geothermal and other emerging technologies face a similar financing gap: capital providers want risk profiles and contract structures that match the project type. The political and market signal from big banks, hyperscalers, and the DOE matters because it reduces perceived stigma around nuclear and encourages more projects to be announced.

Data Points: Stock move in Constellation after the deal: up 30% in a day - Tracy notes Constellation’s share price surged after the Microsoft/Three Mile Island announcement. Three Mile Island reactor closure year: 2019 - Jigar Shah notes one reactor at the site had remained in operation until relatively recently. Loan Programs Office lending authority after IRA: about $40 billion to over $200 billion - Jigar describes how the IRA expanded DOE Loan Programs Office capacity. New load from data centers: at least 25,000 megawatts by the end of the decade - Jigar estimates data centers/cloud growth as a major source of electricity demand. Manufacturing facilities announced: 800 new facilities - Jigar cites reshoring and onshoring as another major load-growth driver. Average nuclear capacity factor: 92% - Used to explain why nuclear production tax credits are lucrative due to high output. Solar capacity factor example: 25% - Compared to nuclear, used to show why nuclear PTC economics can be superior. Restart cost estimate for a nuclear plant: $1 to $2 billion - Jigar gives a rough estimate for restarting a reactor versus building new. Production tax credit value: about 3 cents per kilowatt hour - He says the technology-neutral PTC can be very lucrative for nuclear restarts. Investment tax credit: 30% - He explains that projects can choose the ITC instead of the PTC. Bonus ITC for energy communities: +10% - Additional credit discussed for projects in energy communities. Bonus ITC for domestic content: +10% - Additional credit discussed for using domestic content. Existing U.S. nuclear sites: 54 sites - DOE study referenced as the basis for expansion potential on existing sites. Existing nuclear capacity at those sites: 65,000 megawatts - Jigar says this is available at current nuclear locations. Potential capacity with SMRs/AP1000s on existing sites: almost 95,000 megawatts - Estimate of what could be built at current sites depending on reactor type. SMR cost range: $3 to $6 billion - Jigar cites approximate cost range for small modular reactors. AP1000 cost range: $10 to $14 billion - He cites this as the approximate cost range for larger reactors like Vogtle units. Vogtle workforce training: 13,000 people - Used to illustrate the importance of retaining skilled labor across multiple projects. Utility load-growth stagnation: roughly 20-30 years - Jigar says utilities have had very slow or no load growth for decades. Geothermal cost-effectiveness today: 6 western states - He says enhanced geothermal is currently cost-effective only in a limited region. Craft worker shortage: 500,000 to 1,000,000 people - Estimate of the shortage of trained craftsmen needed for infrastructure buildout.

Pivotal Quotes: "We are going to build big things here" — Jigar Shah: He uses this to describe the policy and cultural shift needed for a nuclear/infrastructure revival. "The first one would cost more, the second would cost less, the third would be even less" — Jigar Shah: He explains the learning-curve economics that are missing when nuclear projects are built one at a time. "If you actually build 10 reactors, it'll be amazing how efficient the NRC will become" — Jigar Shah: He argues regulatory efficiency improves when agencies process more projects and gain repetition.

Implications: If demand growth persists and repeat-build financing works, nuclear could re-enter the U.S. energy mix as a bankable clean-firm resource. The bigger story is that policy, capital markets, and standardized execution may finally align.

🔓 Sign Up for Unlimited Episode Search

About Odd Lots

Bloomberg's Joe Weisenthal and Tracy Alloway analyze the weird patterns, the complex issues and the newest market crazes. Join the conversation every Tuesday and Thursday for interviews with the most interesting minds in finance, economics and markets.

View all episodes from Odd Lots