GAO told the Coast Guard to finish designs first. It refused -- twice
Summary
The Coast Guard's Offshore Patrol Cutter (OPC) program is meant to deliver 25 new cutters, in three stages, to replace its aging medium-endurance fleet. In June 2023, GAO found the program building ships concurrently with an unstable design -- the same practice GAO had first flagged in October 2020 -- while the total cost estimate had already grown from $12.5 billion (2012) to $17.6 billion (2022). GAO recommended the stage 2 program complete its design before starting construction; DHS refused. By November 2025, GAO's follow-up found the predicted outcome: zero ships delivered from either shipbuilder, the stage 1 shipbuilder's contract for two of its four ships terminated for default after costly rework, the lead stage 1 ship now expected more than five years late, the total cost estimate up again to $19.6 billion, and -- despite everything -- construction of the stage 2 lead ship starting in August 2024 without a stable design. GAO recommended, again, that the program complete its design before authorizing more construction. DHS refused, again.
A practice GAO had already warned about once
In October 2020, found the OPC program had started construction on its lead ship with an unstable design. By 2023, nothing had changed: the Coast Guard was still progressing through technology development, design, and construction phases concurrently -- an approach 's own leading practices identify as high-risk, since problems discovered mid-construction require costly rework instead of a design fix on paper. Over that same period, the program's total acquisition cost estimate had grown from $12.5 billion in 2012 to $17.6 billion in 2022, and the first four ships had already slipped 1.5 years due to Hurricane Michael and propulsion-system manufacturing issues.
View data as table
| 2012 estimate | 12.5 |
|---|---|
| 2022 estimate | 17.6 |
| 2023 estimate | 19.6 |
The fix GAO asked for, and the refusal
made five recommendations in 2023. Two mattered most: that the stage 2 program demonstrate all its critical technologies in a realistic environment by the design-review stage, and that stage 2 achieve a genuinely stable design -- full completion of basic and functional design, including the ship's distributive systems -- before construction of the lead ship began. concurred with three of the five recommendations. It rejected those two. 's response was blunt: the Coast Guard was "proceeding down the same risky path as OPC stage 1," which had already suffered delays from exactly this kind of premature construction start.
Two years later: zero ships, one contract terminated
's November 2025 report found that neither shipbuilder -- stage 1's or stage 2's -- had delivered a single ship. Stage 1's problems had compounded: the Coast Guard issued stop-work orders on OPC 4 in March 2025 and OPC 3 in May 2025, then terminated the OPC 3 and 4 portions of the contract for default in July 2025, without yet deciding how those ships would be completed. The lead ship, OPC 1, was down to a single shipbuilder estimate of December 2026 at the earliest -- more than five years later than the program's original delivery goal. The total cost estimate had grown again, to $19.6 billion: 57% above the 2012 estimate, and 11% above just the 2022 figure 's prior report had cited.
View data as table
| Total OPCs planned | 25 |
|---|---|
| OPCs delivered to date | 0 |
The same mistake, with the second shipbuilder
Stage 2, under a different shipbuilder, had done some things right -- collaborative design reviews that supported timely decisions, among them. But construction of OPC 5, the stage 2 lead ship, began in August 2024 without a stable design -- authorized without even meeting the design-maturity parameters in the Coast Guard's own standard operating procedure. 's new recommendation asked the program to demonstrate a stable stage 2 design, including completing the routing of major distributive systems and successfully testing the davit in a realistic environment, before authorizing construction of any more stage 2 ships. rejected it, arguing its existing procedure already covered it. disagreed: that procedure, it noted, was the same one in place when OPC 5's construction started without meeting its own standards, and 's original October 2020 recommendation on this exact point remains open.
View data as table
| Concurred with (3 in 2023, 2 in 2025) | 5 |
|---|---|
| Not concurred with (2 in 2023, 2 in 2025) | 4 |
The takeaway
- 's 2023 warning was specific, and it was ignored. rejected the two recommendations that would have required a stable design before construction -- the exact practice whose absence had already delayed stage 1.
- The predicted outcome arrived on schedule. By 2025: zero ships delivered, two of stage 1's four ships terminated for default, the lead ship over five years late, and the total program cost up to $19.6 billion.
- The mistake wasn't a one-time failure -- it repeated with a different shipbuilder. Stage 2's lead ship, OPC 5, started construction in August 2024 without a stable design, the same condition flagged on stage 1 in 2023. rejected the follow-up recommendation to fix it, too.
Findings on the OPC program's concurrent design/construction practice, its 2012-2022 cost growth, and 's 2023 response are from -23-105805, 'Coast Guard Acquisitions: Offshore Patrol Cutter Program Needs to Mature Technology and Design' (June 2023, revised August 1, 2023 for a construction-percentage correction only), read directly in full via an archived copy after the current gao.gov asset URL blocked direct access. Findings on zero ships delivered, the stage 1 contract termination, OPC 1's revised delivery estimate, the 2023 cost update, OPC 5's unstable-design construction start, and 's 2025 response are from -26-107583, 'Offshore Patrol Cutter: Coast Guard Should Gain Key Knowledge Before Buying More Ships' (November 25, 2025), also read directly in full via an archived copy. This report's OPC-specific cost figures differ from those in 's separate, broader annual assessment of 's major acquisition programs (-25-107317, covered by a separate BlackLeaf article on acquisition-baseline documentation practices) because that report converts costs into constant fiscal-year dollars, excluding inflation, while this report uses then-year dollars -- a -noted methodological difference, not a discrepancy. The two OPC-specific reports examine the same program about 2.4 years apart -- the first warning that skipping a stable design before construction was an avoidable risk, the second confirming that the risk materialized on stage 1 and was then repeated on stage 2.
Sources(2) ▾
- U.S. Government Accountability Office, Coast Guard Acquisitions: Offshore Patrol Cutter Program Needs to Mature Technology and Design (2023-06-01) — -23-105805 (revised August 1, 2023 to correct a construction-completion percentage on page 37; no other substantive change), a report finding the Coast Guard's Offshore Patrol Cutter (OPC) program was building its lead ship concurrently with an unstable design, contrary to shipbuilding leading practices -- the warning facet. Distinct from doc-gao26-107583 (the confirmed-outcome facet, published about 2.4 years later). This report is separate from, and uses different cost-accounting conventions than, 's -wide annual major-acquisitions assessments (e.g. -25-107317, used by a separate BlackLeaf article on acquisition-baseline documentation); the OPC-specific cost figures here differ from those in that other report because of different dollar-year conversions, as itself notes in doc-gao26-107583. Direct gao.gov PDF fetch returns HTTP 403; fetched in full via an archived copy. gao.gov · original document
- U.S. Government Accountability Office, Offshore Patrol Cutter: Coast Guard Should Gain Key Knowledge Before Buying More Ships (2025-11-25) — -26-107583, a report to the House Transportation and Infrastructure Committee finding zero OPCs delivered, the stage 1 shipbuilder's OPC 3 and 4 contracts terminated for default, and the stage 2 shipbuilder repeating the same unstable-design construction start flagged in 2023 -- the confirmed-outcome facet. Direct gao.gov PDF fetch returns HTTP 403; fetched in full via the files.gao.gov HTML mirror (HTTP 200). gao.gov · original document
Comments
Always open. Logged-in readers can annotate paragraphs in place.
The Coast Guard's Offshore Patrol Cutter (OPC) program is meant to deliver 25 new cutters, built in three stages by different shipbuilders, to replace its aging medium-endurance fleet. In June 2023⧉, found the program repeating a mistake it had first flagged in 2020. In November 2025⧉ -- about 2.4 years later -- reported on what came of that, and found the mistake repeating again, with a different shipbuilder.