DOE Eases Review of $65.5B in National-Lab Construction
Summary
In March 2025, the Secretary of Energy ordered DOE and the National Nuclear Security Administration to streamline how they manage the $65.5 billion in construction and acquisition projects underway at national laboratories and nuclear sites run by outside contractors -- delegating more approval authority to those contractors and scaling back the independent reviews meant to catch cost and schedule problems early. As of January 2026, 66 of the agencies' 80 capital projects could be touched by the change. DOE's Office of Science has already handed lab directors -- who are contractor employees, not federal officials -- approval authority over 20 projects at nine laboratories, even though the formal order revision itself remains on hold. GAO's own review found the rollback could leave the agencies less confident in their cost and schedule estimates and slower to catch problems -- and that neither DOE nor NNSA has set goals or measures to tell whether the streamlining is actually working.
A senior signature, a lower bar
Order 413.3B has governed any capital project over $50 million for years, requiring progressively more senior officials to sign off as a project clears each cost-dependent "critical decision" checkpoint -- design approval, construction start, and so on. The March 2025 memo didn't touch that $50 million floor. It inserted two new thresholds inside the range above it: contractors now get more of the sign-off authority themselves for projects up to $300 million, and projects between $300 million and $1 billion get fewer of the independent, outside-the-contractor reviews that used to be standard. It's a narrow-sounding change in a technical order. It applies to a portfolio worth as much as the annual budget of a mid-sized federal agency.
View data as table
| Total capital projects (Jan. 2026) | 80 | collectively estimated at up to $65.5 billion |
|---|---|---|
| Could be touched by the 413.3B revision | 66 | 82.5% of the portfolio, plus future projects |
| Already delegated to contractor lab directors | 20 | Office of Science alone, across 9 laboratories |
Implemented before it was official
As of March 2026 -- a full year after the Secretary's memo -- officials told the formal revision to Order 413.3B was still on hold. That didn't stop parts of the department from acting on it anyway. 's Office of Science had already delegated critical-decision approval authority to national laboratory directors -- who are M&O contractor employees, not federal officials -- for 20 projects across nine laboratories. , separately, was still developing its own guidance on how to scale back independent reviews. 's own framing is blunt about what that means in practice: the changes are "too recent to have yet demonstrated an effect" either way, which cuts both ways -- no proof of savings, and no proof of harm, on a portfolio where a third of the projects already have contractor-employed lab directors approving their own work.
View data as table
| 413.3B applies above this cost | 50 | unchanged: any project over $50M |
|---|---|---|
| New contractor delegation ceiling | 300 | more approval authority now sits with contractors, up to this cost |
| Reduced independent review, upper bound | 1,000 | independent review is scaled back for projects in the $300M-$1B band |
What an independent review actually catches
and officials told that fewer independent reviews would improve efficiency and save money. didn't take that on faith. Its own example of what the old process is for: a 2026 peer review -- one of the independent checks now being scaled back for mid-size projects -- "helped support planning efforts to address expected cost increases" on an project in New Mexico. doesn't say the review discovered that cost growth, only that it fed the planning meant to manage it -- exactly the outside input says is now at risk for any project between $300 million and $1 billion.
No way yet to know if it's working
Asked how they'd know whether the streamlining actually delivers efficiency and savings rather than just fewer eyes on the money, Science and officials pointed to their existing process for evaluating individual M&O contractors' performance. found that answer doesn't hold: contractor performance evaluations measure one contractor at a time, not whether the agencywide policy shift is achieving anything. There is, as of this report, no agencywide goal, outcome, or performance measure attached to a change already touching 66 of 80 active projects and $65.5 billion. recommended 's Secretary and 's Administrator each establish one. Both agencies concurred. Neither recommendation has a completion date -- 's own tracking note reads only that it will "provide updated information" once it confirms what action, if any, the agencies have taken.
- A March 2025 memo loosened contractor oversight on a $65.5 billion, 80-project portfolio -- more delegated approval up to $300 million, fewer independent reviews from $300 million to $1 billion -- and estimates 66 of the 80 projects could be touched.
- Parts of the change were already in effect before the order was formally revised. 's Office of Science delegated approval authority to contractor-employed lab directors for 20 projects at nine laboratories while the order revision itself sat on hold as of March 2026.
- 's own example of what independent review is for: a 2026 peer review -- the kind of check now being scaled back -- helped support planning to address expected cost increases on an project in New Mexico, the exact function says is now at risk for projects in the $300M-$1B band.
- Neither nor has a way to measure whether any of this is working. The agencies' fallback -- evaluating individual contractors' performance -- doesn't assess the agencywide effect of the policy change itself, per 's finding. Both recommended fixes are open, concurred with, and undated.
- This is a forward-looking risk finding, not a report of overruns already happening (high confidence in what states, moderate confidence in eventual impact): explicitly says the changes are too recent to have demonstrated an effect, and this report makes no claim that cost or schedule problems on any of the 66 affected projects have gone undetected yet.
's report covers the general policy shift and its oversight implications; it does not itemize the $65.5 billion across all 80 individual projects, nor does it report project-by-project cost or schedule figures the way 's separate, periodic reviews of 's major weapons-production construction projects do -- this piece draws only on -26-108520 and does not merge its figures with those separate reports. "As much as $65.5 billion" and the 66-of-80 figure are 's own portfolio-wide estimates as stated in the report; the 82.5% share and the 14-project remainder in the chart above are this publication's own arithmetic on 's published counts, not numbers states directly. The order revision's on-hold status and the Office of Science's 20-project delegation are both 's own point-in-time findings (March 2026 and undated within the report, respectively) and may have changed since publication.
Sources(1) ▾
- U.S. Government Accountability Office, Department of Energy: Efforts to Streamline Project Management at National Laboratories and Nuclear Sites (GAO-26-108520) (2026-07-16) — 's July 2026 report examining how and are implementing a March 2025 Secretary of Energy memorandum that directs revisions to Order 413.3B -- the rule governing oversight of capital construction and acquisition projects at national laboratories and nuclear sites run by outside management-and-operating (M&O) contractors. Covers the scope of the portfolio affected, which offices have already acted on the memo despite the formal order revision being on hold, 's own risk finding about reduced independent review, and 's two open recommendations (to and to ) that neither agency has yet satisfied. Fetched directly from gao.gov's Highlights/full-text page (no bot-challenge encountered on this fetch) and independently corroborated against the same content in Artemis's sealed copy. gao.gov · original document
Comments
Always open. Logged-in readers can annotate paragraphs in place.
In March 2025, the Secretary of Energy directed revisions⧉ to Order 413.3B -- the rule governing how the Department of Energy and the National Nuclear Security Administration () oversee construction and acquisition at national laboratories and nuclear sites run day-to-day by outside management-and-operating (M&O) contractors. The stated goal was to cut administrative burden. The mechanism was to (1) let contractors handle more project approvals themselves, for projects up to $300 million, and (2) scale back the independent reviews -- outside checks meant to catch cost and schedule trouble before it compounds -- for projects between $300 million and $1 billion. GAO's own analysis⧉ of the change, released July 16, 2026, found it "could reduce confidence in the reliability of cost and schedule estimates and result in late identification of potential problems." As of January 2026, and were managing 80 capital projects worth up to $65.5 billion combined; the revision could touch 66 of them.