In January, NIA published the report, Right-Sizing Reactors, to explore the economic trade-offs between large-scale reactors, small modular reactors (SMRs) and microreactors. We concluded that both large and small reactors have pathways to lower costs — but they need different enabling conditions to get there, and it's worth being more specific about the policies each technology class needs to accelerate deployment.
Large reactors like the AP1000 — South Korea and China have both delivered real cost declines building at scale — but only under a narrow set of conditions:
-
Strong, sustained demand growth, so a utility or country has a reason to keep building
-
Patient, well-capitalized owners (e.g., state-owned or state-supported utilities, or utilities in regulated markets)Paced, serial deployment rather than one-off builds. Building one AP1000 and stopping is close to the worst-case outcome, a pattern some researchers call "eternal beginner syndrome," where a project type never acquires enough experience to capture learning effects. For example, Vogtle Unit 4 had lower construction costs than Unit 3, partially from learning.
-
A coordinated industrial policy and a robust domestic supply chain, particularly for large forgings, which only a handful of facilities worldwide can produce
-
Modular fabrication of components to reduce on-site construction.
-
Best-practice project management, since typical cost-overruns at recent large nuclear projects can erase whatever scale economies exist on paper
The case for SMRs and microreactors is more hypothetical but strongly supported by evidence from other energy technologies that are modularly produced. However, SMRs need a different set of enabling conditions to succeed; government, industry, and civil society are working to put them in place, but they aren't fully in place yet:
-
High-volume licensing at the NRC
-
Genuine factory fabrication, not just modular components assembled on-site, to capture the benefits of learning-by-doing .
-
A committed order book, since the entire economic case rests on getting from the first to the nth-of-a-kind.
-
Minimal on-site construction and labor, since low on-site labor productivity is one of the most reliable predictors of cost overruns in nuclear projects historically
-
Minimal on-site staffing, both to control operating costs and to fit the off-grid and remote markets where many microreactors are aimed first
Turning these into policy recommendations, three levers matter most for early-mover projects of any size:
Financing. Large projects need access to low-cost capital structures like the one the U.S. Department of Energy Office of Energy Dominance Financing (EDF) uses, along with vehicles like the ARC Act, international investment, and government equity investment to de-risk the first movers. Because they are first-of-a-kind, SMR and microreactor developers need financing that can absorb early-mover risk without requiring an order book to already exist — meaning public support is essential to solve this chicken-and-egg problem .
Demand-side policy. Market-pull policies, government procurement, tax credits, and clean energy portfolio standards help create the willing customers that all size classes need — but they matter differently: for large reactors, they help sustain the paced deployment that avoids eternal beginner syndrome; for SMRs and microreactors, they help build the order book that unlocks factory fabrication in the first place.
NRC reform. Large reactors mostly need continued, predictable licensing for known designs. SMRs and microreactors need something more structural: clearer guidelines for manufacturing licenses, and a licensing process efficient enough to handle a much higher volume of reviews without each one taking years.
In the last few years, there has been significant progress for deploying new nuclear technologies, but some key policy needs remain. This year, EDF announced a $17.5 billion commitment to finance advance purchases of long-lead-time components for up to 10 Westinghouse AP1000 units, which in NIA’s view, is exactly the type of policy that’s needed. However, the Washington Post reported the following month that none of the seven prospective utility partners would identify themselves and no orders for AP1000s have been announced — an indication that early-mover risks still need to be addressed, in addition to the supply chain. We saw that cost declined between Vogtle-3 and Vogtle-4, and Westinghouse got the updated AP1000 design certification approved by the NRC, so the next builders can benefit from all the improvements. But if it takes several years for the next AP1000 project to start construction, some of the key experience may be lost.
The re-introduced Accelerating Reliable Capacity (ARC) Act, which represents one approach to addressing early-mover risk, has been sitting in committee since February. The bill would create a federal risk-reduction program, providing project completion insurance and enhanced financing terms for at least three advanced nuclear projects, authorized at $3.6 billion.
And the Export-Import Bank, the U.S. government agency that funds export projects of American goods, needs reauthorization before it expires at the end of 2026. ExIm has already signed agreements to finance nuclear projects in Romania and Poland. It will be even more critical going forward to help nuclear vendors fill out their order books and to support exports of U.S. nuclear technology, which has benefits for U.S. non-proliferation policy and diplomatic objectives in addition to competitiveness and economic value.
The Nuclear Regulatory Commission has made notable progress with reducing timelines and providing clearer applicant expectations. However, the implementation of the new rules that were directed by Executive Order 14300 could result in issues or delays. Many of these rules will need further adjustments once applicants actually start applying under them, and bottlenecks are uncovered. The NRC will need to actively engage with industry and stakeholders on the issues to efficiently find solutions that are workable. Additionally, the NRC faces a challenge with retaining experienced staff. Some applicants have experienced issues with NRC staff being reassigned from their team to cover other areas, requiring the applicant to refamiliarize the staff with their technology (See NIA’s Licensing Efficiency 2026 Update for further challenges and recommendations).
Regarding high-volume licensing, NIA’s 2024 report, Enabling High-Volume Licensing, laid out three recommendations:
-
Increase the use of standardized new reactor applications and leverage existing regulatory tools to minimize the scope of new or site-specific safety reviews
-
Enable use of alternative environmental review processes that scale staff effort and public review based on the expected and demonstrated environmental impact of new reactor projects.
-
Enable the Commission to use less time- and resource-intensive oversight processes and eliminate the requirement for the NRC mandatory hearing in Section 189a of Atomic Energy Act.
NRC reform is underway on all three fronts such as the proposed Part 57 rule and other ADVANCE Act reforms, but NRC must complete its work efficiently while maintaining public trust.
Across the board, what’s needed most is a combination of market-pull policies and financial risk mitigation measures for early movers. In the long run, new nuclear plants will be more reliable, resilient, and less polluting. And with the right policies the technology can also be affordable and faster to deploy. A federal commitment to new nuclear, including more policy support for financial risk mitigation, is essential for utilities to make such a large upfront investment. SMRs still need public-private partnerships for early movers, but they have a slight advantage over larger reactors because they can more readily find private financing for niche markets like industrial co-location and behind-the-meter data centers. However, building out order books in the tens to hundreds will require broader and stable demand-side policies.
Read the full report: Right-Sizing Reactors: Balancing trade-offs between economies of scale and volume.