Piece of (Yellow)Cake - The Easier Parts of the NRC-Reform Rule

The NRC is under pressure to meet new executive mandates for speed, accountability, and realism. Even the most straightforward aspects of the executive order will likely lead to fights in rulemaking space given the brevity of the direction.

I’ve written previously about the overall pace and content of the major rulemaking that the NRC must conduct under a recent executive order. Given another executive order that adds more layers of White House review, it’s not likely that we’ll have a feel for where the NRC is headed for a while. In the meantime, I’m sharing my thoughts on which parts of the rule can be done in a somewhat simple manner—with, of course, plenty of devil in the details.

Deadlines

The EO directs the NRC to establish binding deadlines for all licensing activities, made concrete through hourly fee caps. The deadlines include a maximum of 18 months for new reactor applications and 1 year for license renewals, with no extensions allowed except for applicant failures.

The NRC will likely try to make this as simple as possible, such as an amendment in 10 CFR Part 2, which governs its licensing procedures. That will effect the change without needing to open a whole bunch of other sections (which would cause other procedural and version-control issues). The fee cap will also need to go into 10 CFR Part 170 somehow—but that’s the only clear part on fees, as noted below.

If they get approval for incorporation by reference, they could just point to their generic milestone schedules as expectations unless applicants fail to perform. Otherwise, they could pick the two actually specified (new licenses and renewals) and add those deadlines in Part 2 and move on.

This direction, however, has a ton of gray areas that will have to be worked out in rulemaking.

“an application to construct and operate”: To me, this only technically addresses a combined license (the only thing that lets you construct and operate). I expect the statement in Part 2 (or wherever it ends up) will be broad enough to cover operating licenses, construction permits, and early site permits, too. People want fixed schedules and anything else will feel like loophole-ing.

“first required step in the regulatory process”: Lots of steps are NOT required but are done by procedure. I expect the kickoff step to be docketing, since acceptance reviews aren’t technically required and this gives some flexibility to start the clock once the NRC knows the review is feasible.

“applicant failure”: How fail-y does the failure need to be? Is a bad first-round RAI response that needs a follow-up a failure? What if the question was vague? I expect the rule to be silent on this, and there to be a few fights before it’s settled. The best analogy is the materials licensing process, which has quite strict deadlines, combined with ways in the licensing system to stop and start a review clock when there are major issues. (See a random example of such a suspension here.)

“shall adopt shorter deadlines … as appropriate”: This is a good-sounding general direction, but I doubt this would make it into a rule because there are no specifics. I expect the NRC guidance to address this as part of the schedule-setting process during the acceptance review.

“fixed caps on the NRC’s recovery of hourly fees”: This, again, sounds great, but how is it actually going to be done? You can’t really enforce a duration (12-18 months) via an integrated amount of hourly fees without knowing how many people are working on the project, how hard. Furthermore, who’s paying when the hourly meter stops running? Right now, anything not recovered by hourly fees (unless specifically relieved/exempted from fees) just gets rolled into the annual fees charged to licensees. I expect the rule to be pretty quiet about how this would be implemented, so that the NRC can deal with it in a future fee rule. It’s possible that the NRC would establish the hours needed during the acceptance review and that the Commission would establish a fee relief category for anything beyond that.

Environmental Reviews

The EO tells the NRC to revise its regs for complying with the National Environmental Policy Act to reflect the 2023 amendments to that statute and the policies articulated in the EO on Unleashing American Energy (mostly about limiting delays in permitting)

The NRC’s report to Congress this January outlined what it was already working on in environmental space, and it’s a lot. Examples include:

* Meeting schedules set out in the Fiscal Responsibility Act

* Reducing page counts

* Making sure reviews are commensurate with the level of potential impact

* Completing consultations in a timely way

* Planning projects better

* Using technology to streamline audits, comment response, and publication

The Commission is also considering rule changes to make environmental reviews simpler, in SECY-24-0046. Rolling those pre-vetted changes into the larger rule would be efficient. However, the Commission has been silent, and there’s a lot of other work going on. It’s probably easier to say that rule changes to meet NEPA aren’t actually needed right now (the statute controls) and wait till a few projects have been done. Then, a later rule could roll up a variety of efficiencies.

Stringent Standards for Changes

The EO also requires “stringent thresholds for circumstances in which the NRC may demand changes to reactor design once construction is underway.” This one puzzles me a bit. The backfit rule (10 CFR 50.109) and parallel “issue finality” requirements in 10 CFR Part 52 already do this. It sets out circumstances when new positions or requirements can be imposed: adequate protection, compliance, or a substantial safety enhancement for which the costs are justified (with a few other noodles in Part 52). Maybe the rulemaking notice can just remind people of this.

What I remember happening at Vogtle was that issues cropped up during construction that hadn’t been contemplated during the licensing review. The design, as licensed, just didn’t work or was built slightly differently. The NRC may have required the licensee to go back and do what the licensee promised, or the licensee may have proposed a license change to move forward as built. This still adds time and uncertainty, and is part of why many vendors are considering the two-part construction + operation process, to not lock in too much up front.

If the goal is really “don’t bother the NRC too much with minor deviations,” then there is probably another way at this. Putting more of the design under the scope of 10 CFR 50.59 (ITAAC, for example) is what immediately jumps to mind, and could be tackled in the rule.

Outside of the rule, this also argues for applicants not trying to get too much resolved in a construction permit. The uncertainty of having them addressed in a future review may be well worth the construction flexibility.

Other Stuff

Revise the Reactor Oversight Process

Likely not much will go in the rule on this point, since most of the ROP lives in policy and inspection documents.

Right-Size Security

The same bullet also calls for changes to “reactor security rules and requirements,” which the NRC could say is already underway in the advanced reactor security rulemaking and take no further action. (Knowing the NRC, though, they’ll feel compelled to do way more.)

Focused Safety Assessments

The EO directs the NRC to ensure that reactor safety assessments focus on credible, realistic risks. Most of the details on safety assessments appear in guidance. A wholesale ripup of what design-basis events are considered in licensing is not likely to happen in this fast-paced rule. Instead, there are two fairly simple ways to tackle this:

* refer to the increased enrichment rulemaking, which already risk-informs loss-of-coolant accident analyses

* take credit for the risk-informed process outlined in Regulatory Guide 1.233; though it’s for non-LWRs, others could apply it with exemptions used where warranted

The Bottom Line

The rulemaking effort has some low-hanging fruit—but expect the NRC to move cautiously, using vague rule text and internal guidance to buy time and flexibility. Even the provisions that seem simple will spark fights, but the intent will be clear and help the industry gain certainty.