Deepwater: the Coast Guard boats that were lengthened u...

In September 2004 the cutter Matagorda came back to port with a problem no one wanted to look at closely. The boat had been cut apart and stretched from 110 feet to 123 feet under a Coast Guard modernization program called Deepwater, and now its hull was deforming. Over the next two years the same failures showed up on seven sister boats. By November 30, 2006, all eight of the lengthened cutters were pulled from service. The hulls had buckled, decks had cracked, propeller shafts had fallen out of alignment, and Coast Guard and Navy engineers could not agree on a single root cause. Roughly $100 million was gone.

That is the sharpest, best-documented piece of waste in the entire Deepwater story, and it is worth separating from the rest at the outset. The 123-foot cutter fiasco is a specific, contained failure: eight boats, one bad conversion, about $100 million. Deepwater as a whole was a $17 billion to $24 billion effort to rebuild almost the entire Coast Guard fleet over two decades. The two numbers do not belong in the same sentence except to say they are different. What connects them is a structural decision about who was in charge, and that decision is where the real lesson sits.

What Deepwater was

The Integrated Deepwater System was awarded in June 2002. According to the U.S. Government Accountability Office (GAO-07-446T), it began as a roughly 20-year, $17 billion program to replace the Coast Guard's aging ships and aircraft. After the September 11 attacks reshaped the service's missions, the plan was re-evaluated. A July 2005 revised implementation plan expanded Deepwater to about 25 years and roughly $24 billion. When you see the two figures, the years matter: $17 billion is the 2002 twenty-year figure, and $24 billion is the 2005 twenty-five-year figure.

The reason the fleet needed replacing was not in dispute. By the early 2000s the Coast Guard was operating some of the oldest ships and aircraft in the U.S. armed forces. Maintenance costs were climbing, availability was falling, and legacy hulls were being asked to do post-9/11 work they were never designed for. Recapitalizing that fleet was a legitimate public need. The ambition to do it as an integrated system, rather than one ship at a time, was defensible in principle.

The delivery mechanism is what made Deepwater unusual, and what the government's own watchdogs later identified as the flaw. The Coast Guard ran the program through a private "system integrator," Integrated Coast Guard Systems (ICGS), a joint venture of Lockheed Martin and Northrop Grumman. ICGS was given broad authority to make design and procurement decisions on the government's behalf. GAO found that the Coast Guard kept too little in-house engineering oversight to check that work, and concluded that the system-integrator model provided "little protection against cost growth."

The 123-foot conversion, step by step

The plan was to convert all 49 of the Coast Guard's 110-foot Island-class patrol boats into 123-foot cutters, extending their hulls and upgrading their sensors and communications. Only eight were done before the program was halted. Per the DHS Office of Inspector General (OIG-07-27, February 2007), every one of those eight failed.

The timeline is tight and well documented:

September 10, 2004: the Matagorda, the first converted cutter, suffers hull deformation. Nunivak, Padre, Metompkin, Vashon, Monhegan, and others follow.
2005: Commandant Admiral Thomas Collins decides to stop the conversion at eight hulls rather than continue toward 49.
November 30, 2006: the Coast Guard suspends operations of all eight 123-foot cutters because of continuing hull deformation, some of it resulting in hull breaches.
2013: the hulls are formally decommissioned, with about $30 million in equipment salvaged, as reported by Military Times.

The DHS Inspector General catalogued the defects: hull buckling and deformation, deck cracking, interior frames bulging and warping, and shaft-alignment problems. Investigators added a second finding that matters as much as the cracks. The contractor's command, control, communications, computers, intelligence, surveillance, and reconnaissance (C4ISR) equipment did not comply with the design standards written into the Deepwater contract. Coast Guard and Navy engineers, working the problem together, could not isolate a single root cause. The boats were found unseaworthy.

The money, kept honest

The commonly cited figure for the 123-foot loss is "roughly $100 million." That number is a rounded headline, and it is worth being precise about what sits underneath it. The Coast Guard reported its own conversion spend at about $87 million. Congressional sources cited about $94 million. A Congressional Research Service product (R42567) and related press more often describe the refurbishment cost at around $88 million. Some broader accounts push total losses as high as roughly $150 million. The clean way to state it: about $100 million, with the documented conversion spend clustering in the high $80s to low $90s of millions. Anyone who quotes a single exact figure is claiming a precision the record does not support.

There is a coda that puts a specific number on the shipyard's role. In 2011 the United States sued Bollinger Shipyards, the yard that performed the lengthening, under the False Claims Act. The Department of Justice alleged that Bollinger had represented the converted hulls' longitudinal strength as roughly twice its actual value, that it ran the strength calculation three times and provided only the highest and most inaccurate result, and that it skipped mandated quality control. The government's suit sought far more, framed in press coverage as a roughly $79 million dispute under treble-damages math. It settled in 2015 for $8.5 million. Report the settlement as $8.5 million; that is what was paid, not the $79 million the government initially pursued.

One clarification on the shipyard, because there is a name collision in the public record. Bollinger reached a separate, unrelated False Claims Act settlement of $1,025,000 in 2025 concerning workers ineligible to work in the United States on Fast Response Cutter contracts. That is a real settlement, but it has nothing to do with the 2000s 123-foot hull-strength case.

It was not only the small boats

The 123-foot cutters are the cleanest failure, but the same weak-oversight model left marks on the larger ships too, and here accuracy cuts both ways.

The National Security Cutter, the fleet's largest new class, had a hull-fatigue design problem. A Navy assessment in August 2006 found areas with insufficient fatigue strength for the expected 30-year service life, with some parts modeled to survive far less. The design was corrected for later hulls, and the first two ships, Bertholf and Waesche, were retrofitted with structural reinforcement during depot maintenance. This was a design shortfall that was caught and fixed, not a catastrophic in-service collapse like the 123s. It should not be overstated.

The original Fast Response Cutter tells a similar story of an early design that did not survive contact with engineering reality. The first Deepwater-design FRC used a composite hull, and it was suspended in 2006 over hull-shape, weight, and feasibility concerns, per GAO-06-764. The Coast Guard then switched to a proven, off-the-shelf parent-craft design, the Sentinel class, which was built in large numbers and is in wide service today. The failure was the original composite design. Today's Fast Response Cutter is a different, later, successful ship, and nothing here should be read to imply otherwise.

Taking the wheel back

Around 2007 the Coast Guard reasserted control over its own acquisitions. Commandant Admiral Thad Allen announced that the service would resume the lead role, consolidating Deepwater under its own Acquisition Directorate, known as CG-9, and moving away from the ICGS lead-system-integrator arrangement, which was ultimately not renewed. The change was phased across roughly 2007 to 2011 rather than accomplished in a single day, but 2007 is the correct anchor for the decision. Note that two different commandants sit at the two turning points: Admiral Collins made the 2005 decision to stop the 123-foot conversion at eight hulls, and Admiral Allen led the 2007 takeback.

The honest failure, and the honest defense

The critique. Deepwater's central efficiency failure was structural, not incidental. The Coast Guard outsourced its own acquisition judgment. It handed a private joint venture broad authority to make design and procurement decisions on the government's behalf while keeping too little in-house engineering oversight to catch bad work before money was spent. The predictable results were eight patrol boats that buckled and were scrapped at a cost near $100 million, a shipyard that later paid $8.5 million after the Justice Department alleged it misrepresented hull strength, a national-security-cutter design that had to be reinforced, and a fast-response-cutter design that had to be thrown out. GAO's recurring conclusion was that the acquisition structure gave "little protection against cost growth" and lacked a mechanism to ensure the assets met their requirements. When the government does not keep enough of its own expertise in the room, it loses the ability to know whether it is buying something that works.

The defense. The mission was sound. The fleet genuinely was decrepit, and rebuilding it was a real public good, not a boondoggle dressed as one. The instinct to modernize as an integrated system was reasonable on paper. And the story has a genuine recovery. After the Coast Guard reclaimed acquisition authority and stood up CG-9, the program delivered. The National Security Cutters were reinforced, built, and put to sea. The redesigned Sentinel-class Fast Response Cutter, built to a proven design instead of an untested one, has been delivered in quantity and does its job. The lesson is not that fleet modernization was wasteful. It is that a government cannot contract out its own oversight and expect to keep control of the outcome.

Both verdicts are true at once. The 123-foot cutters were a clear, documented waste of roughly $100 million and a fair symbol of what the system-integrator model produced. And the underlying recapitalization, once the Coast Guard put its own hands back on the wheel, largely worked. The failure was in the structure, not the ambition.

Fact-check notes and sources

Deepwater awarded June 2002 as a roughly 20-year, $17 billion program; expanded in a July 2005 revised plan to about 25 years and roughly $24 billion. U.S. Government Accountability Office, GAO-07-446T. Attach the year to the figure: $17B is 2002/20-year; $24B is 2005/25-year.
Program run through a private system integrator, Integrated Coast Guard Systems, a Lockheed Martin and Northrop Grumman joint venture with broad design and procurement authority; GAO faulted weak Coast Guard oversight and the model's "little protection against cost growth." GAO-07-446T.
Eight 110-foot Island-class boats lengthened to 123 feet; all eight developed hull buckling, deck cracking, deformation with some breaches, and shaft misalignment; no single root cause isolated; contractor C4ISR equipment did not meet Deepwater design standards; all eight suspended November 30, 2006. DHS Office of Inspector General, OIG-07-27 (February 2007).
Roughly $100 million cited loss; documented conversion spend clusters near $87 million (Coast Guard), about $94 million (congressional sources), and about $88 million refurbishment per Congressional Research Service R42567; some accounts cite up to about $150 million in total losses. The $100 million figure is a rounded headline. General framing also in CBS News, The Troubled Waters of Deepwater.
Formal decommissioning of the 123-foot hulls in 2013, with about $30 million in equipment salvaged. Military Times, Coast Guard to decommission troubled 123s.
United States sued Bollinger Shipyards over the failed lengthening; DOJ alleged longitudinal strength represented at roughly twice actual value, three calculation runs with only the highest reported, and skipped quality control; settled 2015 for $8.5 million. U.S. Department of Justice, Bollinger Shipyards agrees to settle False Claims Act suit. The separate 2025 Bollinger $1,025,000 settlement over ineligible workers is unrelated.
National Security Cutter fatigue-life shortfall identified by a 2006 Navy assessment; first two hulls reinforced, design corrected for later ships. Coast Guard and independent engineering reviews; context in GAO, GAO-06-764.
Original composite-hull Fast Response Cutter halted over design problems; Coast Guard switched to the proven Sentinel-class design now in service. GAO-06-764.
Coast Guard reclaimed acquisition authority beginning in 2007 under Commandant Admiral Thad Allen, consolidating Deepwater under its Acquisition Directorate (CG-9); phased across 2007 to 2011. U.S. Senate Committee on Commerce, Science and Transportation, February 2007 oversight hearing; background in CRS, Deepwater program history.

Deepwater: the Coast Guard boats that were lengthened until they buckled