U.S. Shipbuilding:
The Record

In fiscal year 1951, U.S. shipyards delivered 14 oceangoing ships, equivalent to three percent of world output. That share held through fiscal year 1965, supported by the Construction Differential Subsidy, a federal program designed to cover the full differential between American and foreign construction costs for ships operating in international trade (subject to a statutory cap of 50% of total vessel cost). Jones Act vessels in domestic coastwise trade were ineligible for CDS. By fiscal year 1966 it declined to 1.8 percent, and from that point on U.S. shipyards would never again account for more than 3 percent of world ship deliveries.

U.S. output measured in deadweight tons tells a superficially different story, peaking around 1977–1978. The Merchant Marine Act of 1970, which extended subsidies to tankers and bulk carriers and authorized over 300 new ships, provided the policy framework for what became the most significant peacetime shipbuilding activity in U.S. history (though only around 200 of the planned ships were ultimately built). That spike was further assisted by the construction of large tankers to carry Alaska crude oil domestically following the opening of the Trans-Alaska Pipeline System (TAPS) in June 1977.

It was a demand-driven, geographically specific surge, not evidence of restored competitiveness. When pipeline-era demand subsided, the underlying output base that remained was thin and captive. Both the number of ship deliveries and deadweight tonnage make clear that, after the mid-1960s, U.S. commercial shipbuilding ceased to be a meaningful participant in the global market.

The UNCTAD rankings show what the U.S. shipbuilding industry has become in the global context. In 2025, American yards delivered 0.03 percent of world gross tonnage — ranking 19th, behind countries including the Netherlands, Vietnam, and Malaysia. The five-year average from 2021 to 2025 was 0.07 percent, confirming this is not an anomalous performance. In 2025, U.S. yards delivered 25,172 gross tons — less than the combined output of Spain and Portugal (31,284 GT), two countries not generally regarded as major shipbuilding nations.

The industry's retreat from international competition is not recent. According to a 1992 USITC report, U.S. yards had not produced a commercial vessel for export, that is, for foreign-flag operation, since 1960. That remains true today.

Since 2000, U.S. yards have averaged fewer than three Jones Act oceangoing ship deliveries per year, and that average is inflated by the mid-2010s tanker surge driven by the fracking boom and a resulting increase in domestic oil transport demand. With the oil export ban still in place until late 2015, domestically produced crude had to move by Jones Act tanker. In most years, the number of oceangoing deliveries is one or two, and sometimes zero. The U.S. Navy’s own fiscal year 2025 shipbuilding plan, as cited in a February 2025 GAO report, acknowledged that U.S. commercial shipbuilding “has experienced a near-total collapse.”

This age profile is a direct consequence of the cost of building a Jones Act ship. Because construction costs run four to five times international equivalents, operators cannot justify ordering replacements at anything close to normal intervals. Instead, they extend service lives well past those of foreign-flag competitors, keeping aging hulls in operation not because they are well-suited to the trade but because the alternative is often economically indefensible.

The data bear this out. As of January 2026, roughly 30 percent of the Jones Act fleet is 25 years or older. One ship — the Coastal Trader, serving western Alaska — dates to 1963. Among the 22 U.S.-built Jones Act vessels that left the fleet between 2016 and 2026, the median exit age was 42 and the youngest exits were 33. The El Faro, which sank in October 2015, was 40 years old at the time of her loss.

Between 2020 and 2023, three old ships left the Jones Act containership/ConRo fleet (the 1971-built Lihue, 1980-built Kauai, and the 1973-built Matsonia) while three new vessels entered (the 2020-built Matsonia, 2022-built George III, and the 2023-built Janet Marie). This drove the median age from 30 years down to 18.5, not because the industry had turned a corner, but because the fleet is so small that a handful of deliveries moves the needle dramatically. Since 2024, with no further additions, the median has resumed its climb, reaching 21.5 years by January 2026.

The pattern is consistent across vessel types. This age profile describes a symptom, the cause of which is documented in the next section.

The most recent Jones Act containership program illustrates the scale of the premium. In late 2022 Matson ordered three Aloha-class containerships from the Hanwha Philadelphia Shipyard that, per Matson's 2025 Annual Report, cost approximately $335 million each ($1.006 billion for all three). James Lightbourn, founder of the shipping financing advisory Cavalier Shipping, told the New York Times that similar ships from an Asian shipyard would cost $75 million at most, less than one-fourth the U.S. price.

The Aloha-class price is not an outlier. In 2017 Pasha Hawaii ordered two 2,525-TEU containerships — the LNG-powered George III and Janet Marie — from Keppel AmFELS in Brownsville, Texas at a contracted price of $200 million each. Both were originally scheduled for delivery in 2020. George III, however, did not arrive until summer 2022, and Janet Marie until summer 2023. By its delivery, George III had cost over $225 million. Meanwhile, in late 2021 Namsung Shipping ordered a conventionally powered 2,500-TEU vessel from a South Korean shipyard for $41 million. That ship, the Starship Neptune, delivered on schedule in October 2023.

The differing propulsion systems have little explanatory power for the price difference. Industry estimates place the LNG propulsion premium at roughly 20–25 percent over conventional diesel newbuilds. Applied to the Starship Neptune price, that implies an LNG-adjusted equivalent of perhaps $50 million per vessel, which is still less than a quarter of what Pasha paid.

For tankers, the economics are similarly punishing. Kinder Morgan's January 2026 investor presentation states that a new Jones Act medium-range tanker costs over $240 million per vessel, with the earliest possible deliveries in 2029. At that capital cost, a term charter rate of more than $115,000 per day would be required to underwrite new construction. Current Jones Act MR tanker charter rates are approximately $90,000 per day. Although far above international equivalents, such rates are nonetheless 28 percent below the threshold to justify new construction.

In comparison, the international newbuilding price of an MR tanker is approximately $51 million, according to Compass Maritime, and $49.5 million per the BRS Group Annual Review 2026. This makes the Jones Act premium approximately five times the international price.

At $50 million for a new MR tanker, retiring a 20-year-old vessel is often economically rational for international operators, as a modern ship offers lower fuel consumption, lower maintenance costs, and higher reliability. At $240 million, the same calculation is more difficult to pencil out, and in a captive market there is less competitive pressure to remove aging tonnage.

The cost premium is structural. World-class shipbuilders specialize and build high volumes for the global market, while U.S. shipbuilders build small volumes for a protected domestic market.

Volume unlocks economies of scale, drives the learning-by-doing that improves productivity, enables investment in productivity-enhancing technology and equipment, and sustains a supplier ecosystem — the specialized steel fabricators, equipment vendors, and subcontractors — without which no yard can function efficiently. Specialization compounds these gains. Yards that build the same vessel type develop repeatable processes, refined designs, and institutional knowledge that one-off producers cannot accumulate.

These are not novel insights. The observation that specialization and volume are key to competitive shipbuilding has been made continuously, by witnesses with direct operational knowledge, for well over a century.

Writing in 1905 for the American Economic Association, economist Royal Meeker noted that where an American yard turned out one steel vessel, an English yard turned out a dozen, many on the same model. This allowed the use of identical parts and cost reductions that no one-off producer could match.

Harvey D. Goulder, a maritime lawyer and counsel to the Lake Carriers' Association, told the Merchant Marine Commission in 1904 that the difficulty with cost at American coast yards "lies very much in the proposition that they are so largely making new things upon new designs." A yard, he pointed out, might build a battleship one year, then a sidewheel excursion steamer, then a tug, then a freight steamer. Each would use a new design, with no accumulated learning to carry forward to the next vessel.

That diagnosis would be repeated in future decades. Homer L. Ferguson, president of Newport News Shipbuilding & Dry Dock Company — one of the country's most capable yards, then as now — told the 1929 Second National Conference on the Merchant Marine that American yards had still been unable to specialize "except on the Lakes." Describing Newport News's own order book of two barges, a yacht, a passenger vessel, and two cruisers all under construction simultaneously, he described the yard as "country practitioners" rather than specialists.

The remedy, Ferguson argued, was straightforward: "the way to get costs down is for one yard to build only one class of vessel and to build that class in sufficient volume to keep going steadily." He acknowledged that foreign yards faced the same volume constraint, but was clear that "the handicap in this country is much more pronounced than abroad, and until we can get volume we cannot cut costs to any appreciable extent."

Fifty years on, M. Lee Rice, president of Ogden Transportation Corp. and chief executive of Avondale Shipyard, testified before Congress in 1979 that the limited domestic market was "one of the very important factors" in U.S. shipbuilding costs. "If you look at my shipyard today, I am building six different kinds of vessels, all at the same time. That creates by its very nature a higher cost than if I were building a good number of one class of vessel."

The economic logic is well documented. A 1975 GAO study found that series production reduces costs through specialized facilities and learning-curve effects, estimating that realizing the full benefits requires building five to ten ships of the same type — a threshold the U.S. domestic market rarely, if ever, reaches.

The same study offered another key observation. It documented that over 1970–73, more than half the output of the eight largest merchant shipbuilders was for export, while more than a third of the tonnage added to their own fleets was foreign-built. In other words, major shipbuilding nations were simultaneously exporting and importing ships.

This is exactly what economic logic and comparative advantage would predict, with countries specializing in the vessel types they build most efficiently and importing the rest. In contrast, the United States exported roughly 1 percent of shipbuilding output and imported almost nothing — the expected result of a country that prohibits the use of foreign-built vessels in its domestic commerce.

By building small volumes of a variety of vessel types, U.S. yards have few opportunities to apply lessons learned during the construction process. A 2002 analysis by the CNA Corporation put hard numbers on the learning curve, or lack thereof. One U.S. shipyard reported that building the first vessel of a type required three times the labor hours of the final three in a series of thirty, a 67 percent reduction. Korean and Japanese yards building fifty commercial ships a year capture these gains continuously, while U.S. yards building one or two have sometimes even experienced reverse learning, with costs on consecutive ships actually increasing.

When a Korean yard begins a new vessel type, it draws on accumulated knowledge from similar recent builds. U.S. yards typically have no such foundation.

This is reflected in a yawning productivity differential. A 2001 study published in SNAME Transactions, the journal of the Society of Naval Architects and Marine Engineers, found that the gap between the best international yards and U.S. commercial yards was a factor of five to seven, measured on an apples-to-apples basis using the internationally accepted Compensated Gross Tonnage productivity metric.

A 2001 national security assessment by the Commerce Department's Bureau of Export Administration reached the same conclusion. It found that U.S. shipbuilding productivity lagged Europe and Japan not because of wage rates — which were actually higher abroad — but because of organizational efficiency. Not only was the U.S. behind, but the gap had widened over the preceding decade. And it has not closed.

Lawmakers who have visited both U.S. and Asian shipyards in recent years have described the contrast in stark terms. Former National Security Advisor and SHIPS for America Act co-sponsor Mike Waltz, after visiting Hanwha and Hyundai facilities in Asia, observed that some U.S. yards "look like [they] haven't changed since the 1930s." A South Korean lawmaker who visited the Philadelphia Shipyard — the yard responsible for the greatest share of Jones Act ship deliveries since 2000 — after Hanwha's acquisition was similarly blunt, stating that "it looks like a shipyard from the 1980s."

"The productivity is practically nil — and technically it's far behind the Japanese."— Constantine Gratsos, president of Victory Carriers, New York Times, July 1980

The 2001 SNAME Transactions study identified the U.S. shipbuilding quandary as a "chicken and egg" situation: adequate productivity cannot be achieved without adequate throughput, and adequate throughput cannot be achieved without adequate productivity.

A 2008 paper in the Journal of Ship Production by naval architect William O. Gray — a former Bethlehem Steel shipbuilder and recipient of SNAME's highest honor — described U.S. large yards as "jacks of all trades, masters of none," combining Navy and commercial work, repair and newbuilding, without the focused repeat-order production that drives efficiency.

The Jones Act bears responsibility here. Ensconced within a small, insulated domestic market, U.S. shipbuilders lack the volume that productivity requires and the pressure of international competition to specialize and innovate.

As MIT Professor Ernst Frankel observed in a 1996 paper in the Journal of Ship Production, "it is a basic finding of economics that government subsidies, aids, protection, and regulation of an industry will cause its productivity to decline." High costs inevitably result.

One need not be steeped in free-market economics to understand the logic. Speaking at the same 1929 conference that Homer Ferguson of Newport News Shipbuilding addressed, Andrew Furuseth, president of the International Seamen's Union, argued that the build requirement had handed American shipyards "a monopoly unlike anything else in any other trade or occupation in the United States." His prescription was blunt: "There is no way in which you can build ships in competition with the world except by abolishing the monopoly."

Shipyards that do not face world-class competition should not be expected to perform at world-class levels.

"You have given to the American shipbuilder a monopoly the like of which is not to be found in any other trade or in any other calling or in any other occupation in the United States."— Andrew Furuseth, President, International Seamen's Union, Proceedings of the Second National Conference on the Merchant Marine, January 23, 1929

The decline of U.S. commercial shipbuilding cannot be reduced to a single cause, with other contributing factors including management failures, workforce gaps, and the legacy of mixing naval and commercial work. Nor is the U.S. story unique. European shipbuilding has also contracted dramatically as the industry consolidated toward East Asia over the past half century. The Jones Act did not produce this outcome alone, and its absence would not have guaranteed a thriving U.S. shipbuilding industry.

Yet it is almost undeniable that the law, by structurally insulating the industry from competitive pressure and guaranteeing a market regardless of cost or quality, is a major contributor to the sector's poor performance. The only open question is how much better condition it would find itself in today had the United States not gone down its protectionist path.

The Jones Act requires that ships serving U.S. domestic trade be built in the United States. To meet this qualification, vessels must be assembled in a U.S. shipyard, with all major components of the hull and superstructure fabricated domestically. Equipment, outfitting, engines, and machinery are excluded. A vessel incorporating foreign-built modules, provided they are installed in a U.S. shipyard, does not lose its Jones Act eligibility. The law says nothing about the origin of the ship's design, its propulsion system, or the components that go into it.

U.S. shipyards take full advantage of this flexibility. The result is that a "U.S.-built" Jones Act ship may be designed abroad, powered by a foreign engine, and assembled with components sourced from overseas, including China.

This reliance on a foreign supply chain is extensive. A Philadelphia Shipyard executive acknowledged at a 2015 MARAD meeting with the Shipbuilders Council of America that, of the more than one million parts and components in a ship, "at least half of them... are going to be imported."

The president of General Dynamics NASSCO was equally direct in comments submitted to MARAD the same year, noting that "NASSCO's supply chain is routed primarily through Busan, Korea."

A dataset compiled by researcher Caleb Petitt and published by the Independent Institute in April 2026 sheds similar light on the widespread use of imports. Examining the Jones Act fleet's 93 oceangoing cargo ships, it found that no Jones Act vessel with an American-made engine had been built since 2000, no vessel with an American propeller since 2010, no vessel with an American anchor since 1998, and not a single vessel in the fleet has American-built generators. Korean suppliers dominated across all four categories, with Chinese-made components also present across all four.

Two specific vessels, the Isla Bella and Perla del Caribe, help illustrate the extent of Korean involvement in American shipbuilding. As noted by a September 2013 Coast Guard determination letter, NASSCO was under contract to purchase "the vessel design, as well as most of the equipment and material necessary to construct each vessel" from a Korean source. The exceptions were "most steel plate, flat bar, weld rod and paint."

The Congressional Research Service, meanwhile, has noted that NASSCO partnered with Daewoo Shipbuilding for design, engineering, and procurement support, with engines, piping, crew quarters, and portions of the hull imported from Korea and assembled in the United States. Shipyard unions have referred to such vessels as "kit ships."

The dependence is not confined to NASSCO. Veteran-class tankers constructed at what was then Aker Philadelphia Shipyard each required approximately 500 shipping containers of materials imported from Busan as well as an additional 25 bulk shipments for large items such as the main engine, propeller, and anchors. The authors of a 2007 paper on the tanker program observed that "the great majority of equipment, outfitting and materials for these vessels comes from Korea."

The Janet Marie is the most recently built Jones Act containership. Delivered in 2023 by the Keppel AmFELS shipyard in Brownsville, Texas, it provides the latest available snapshot of the U.S. shipbuilding industry's heavy reliance on imports. The anchor chains come from Jiangsu province, and the cranes were manufactured by the China State Shipbuilding Corporation. The propeller and propeller shaft are also Chinese. The propulsion control system is South Korean, as is the main engine, built to a Danish-German design. The generators are Czech. This reflects only what can be gleaned from the major components listed in the Janet Marie's ABS classification record.

There is another foreign dimension. At the time of the Janet Marie's delivery, Keppel AmFELS was owned by Keppel Corporation of Singapore and has since been sold to Turkish firm Karpowership. Other U.S. shipyards are also foreign-owned: Hanwha Philly Shipyard has South Korean ownership, Austal USA in Mobile, Alabama is a subsidiary of Australian firm Austal, and Bay Shipbuilding in Wisconsin is owned by Italy's Fincantieri.

Beyond materials and ownership, U.S. shipyards also use foreign labor. At a 2023 maritime industry forum in Washington, D.C., John Graykowski, a maritime consultant and former Senior Vice President and General Counsel of Aker Philadelphia Shipyard, acknowledged during Q&A that the yard has flown in Korean technicians to perform engine installations. A Korean technician at the yard corroborated the broader pattern, describing in a 2026 account reported by the Chosun Ilbo how Korean experts had participated in shipbuilding there for five to ten years prior to Hanwha's acquisition.

West Coast shipyard NASSCO, which builds both Navy and commercial vessels, employed 36 Mexican engineers to install power plants, engines, and machinery aboard a Navy vessel.

Vessels employed under the Jones Act may carry the designation of "U.S.-built," but the law does not free the United States from foreign reliance for its commercial shipbuilding needs. What it actually produces is something more modest: vessels assembled at U.S. facilities using foreign designs, foreign components, foreign capital, and in some cases foreign labor, all at prices that can run four to five times the international market rate. If there is something distinctively American about a Jones Act ship, it may be less the vessel itself than the price paid for it.

COSCO Nantong Shipyard is a subsidiary of China COSCO Shipping, the state-owned conglomerate that Beijing has cultivated as a cornerstone of its maritime ambitions. It is also, for the past two decades, a preferred maintenance and conversion facility for the Jones Act fleet.

The relationship dates to at least 2001, but gained public attention in 2006–07 when Matson had work done on the Mokihana at COSCO Nantong as part of a conversion to a ConRo vessel. This prompted a federal court challenge in which the Shipbuilders Council of America and Pasha Hawaii Transport Lines, Matson's sole Jones Act competitor in the West Coast–Hawaii trade lane, argued the ships had been so substantially rebuilt abroad as to lose Jones Act eligibility (ships lose their coastwise endorsement if structural steel changes exceed 7.5% of hull weight, with USCG discretion up to 10%). A federal judge, however, sided with Matson.

By 2007 TradeWinds reported that COSCO Nantong had made “something of a speciality out of Jones Act work.” In 2019, Matson and COSCO Nantong held a ceremony to mark the 50th Matson vessel repair, and two years later the total reached 60. In 2023–2024, Matson sent three more Jones Act containerships to COSCO Nantong to retrofit their engines to LNG. The bill totaled approximately $166 million, an amount sufficient to purchase three 2,700 TEU containerships from a Japanese or Korean shipyard with $10 million left over.

This dynamic is not limited to Matson. Pasha Hawaii, which had opposed Matson's early use of COSCO Nantong in federal court, subsequently elected to have substantial work performed on its own vessels at the same yard. In both 2021 and 2025, Pasha received written confirmation from the U.S. Coast Guard that converting its 1980-built containerships Horizon Reliance and Horizon Spirit from steam power to LNG at COSCO Nantong would not affect the vessels' Jones Act compliance. The conversion of the Horizon Reliance has been completed, with Pasha touting the ship — newly renamed George II — as the world's first steam-to-LNG conversion.

That it was the first vessel in the world to undergo this conversion is due to the unique economics of the Jones Act, in which repowering ships well past 40 years of age to LNG — which carries a price tag in the tens of millions of dollars (Matson paid $72 million to retrofit one of its ships, the Manukai, to LNG power) — is more economically rational than purchasing a new U.S.-built ship.

The role of Chinese shipyards in Jones Act vessel repairs and conversions extends beyond the Pacific trade lanes. National Shipping of America's National Glory, which serves Puerto Rico, was converted from a general cargo ship to a containership at the Chengxi Shipyard in Jiangsu province. Seabulk Tankers sent two Jones Act tankers to China for double-hull conversion after finding the all-in cost was roughly one-third of the best American quote. Seacor chairman Charles Fabrikant said he was “shocked when I saw the numbers.”

The cost advantage over U.S. yards is even more notable after considering that foreign vessel repairs are subject to a U.S. duty of 50 percent. Despite this, a MARAD-commissioned survey found that carriers reported using foreign shipyards for American-flag ship repairs because “the cost of having repairs performed overseas and paying the duty was often lower than having the repairs performed in U.S. shipyards.”

But U.S. shipyards' lack of cost competitiveness is not the only driver of foreign repairs and overhauls. Availability is another issue. Jared Gale of Dole testified at a 2025 USTR Section 301 hearing that when the company sought U.S. drydocking for two vessels, one yard never responded, one declined while waiting on a Navy contract, and a third lacked the necessary skills. Dole had no choice but to send the ships overseas. In July 2017, Resolve Marine received USCG approval to drydock its vessel Resolve Pioneer in Zhoushan because no suitable drydock was available in Alaska.

The relationship between Jones Act carriers and Chinese shipyards is a study in unintended effects. In theory, the Jones Act ensures a steady demand for U.S.-built ships to sustain American shipyards and promote national security. In reality, U.S. shipyard costs have become so out of step with international norms that aging ships are dispatched to the repair facilities of a leading geopolitical rival to extend their working lives far beyond what U.S. economics would otherwise allow.

The decline of American commercial shipbuilding is not a recent phenomenon. The industry has not been internationally competitive since the era of wooden ships, and its extraordinary performance during World War II only temporarily obscured a long-standing competitive deficit. The postwar decline is properly viewed not as the fall of a shipbuilding giant, but as a logical regression to an uncompetitive baseline. The country's shipbuilding problems are not new but long-standing and well-documented.

That this has taken place under the Jones Act's watch is neither surprising nor coincidental. The law is not rooted in timeless principles, but in the conditions of a particular moment, when American shipbuilders were genuinely competitive and the difference between a merchant vessel and a warship was far less pronounced. In other words, a world far removed from today's. It is both notable and instructive that no other shipbuilding country of significance relies on such severe protectionism to advance the fortunes of its industry.

American shipbuilding in the country's early days was a highly competitive industry, so much so that when the Revolution began, an estimated one-third of all seagoing merchant ships flying the British flag were of American construction. Vast forests and ample supplies of timber, along with skilled shipwrights, meant that, as Winthrop Marvin noted, a live oak vessel could be built in New England for $38 a ton compared to at least $50 per ton in Britain. Protectionist measures to encourage the use of U.S.-built ships therefore imposed little if any cost.

According to maritime historian Winthrop L. Marvin, the requirement that American ships be built in American yards was "by no means an actual handicap," with such vessels so competitively built that "there was generally no advantage" in purchasing them abroad. The Abandoned Ocean by Andrew Gibson and Arthur Donovan concurs, declaring that early forms of protection "cost the nation nothing, since the American producers of these goods (ships) and services (shipping) were highly competitive."

This advantage, however, would dissipate as iron replaced wood and steam replaced sail. According to F.G. Fassett Jr.'s authoritative 1948 history, The Shipbuilding Business in the United States of America, Great Britain was already building 50 percent of its vessels in iron or steel as early as 1864. In contrast, the United States did not reach that threshold until 1900. U.S. shipyards lost their cost advantage shortly before the Civil War, and by 1869 U.S.-built wooden ships cost about as much as British iron ships. Rather than innovate, Gibson and Donovan observed that American yards "stuck to what they did best, building and operating wooden sailing ships." This is exactly the type of behavior one should expect in a market insulated from foreign competition.

As the steam-and-steel transition became entrenched, U.S.-flag shipping operators sought permission to purchase foreign-built steamships. The shipbuilding industry, still dominated by wooden vessel construction, strongly opposed the measure, and it failed to gain sufficient political support. In the Naval War College Review, Christopher McMahon notes that “as a result, many U.S.-flag shipping companies went out of business.”

Once lost in the 1860s, American cost competitiveness in shipbuilding was never regained. The American cost premium has been consistently documented by government bodies and industry observers ever since:

Both world wars saw enormous surges in U.S. shipbuilding output. These spikes, and particularly that of World War II, have fed a narrative that the U.S. is now a shipbuilding giant brought low, and that its wartime performance reflected its underlying commercial prowess. But this misreads history. While the production increases were genuine achievements of industrial mobilization, they were not evidence of peacetime commercial competitiveness. Nor were they the fruit of the shipbuilding protections offered by U.S. coastwise laws.

The U.S. experience in World War I laid bare the inadequacy of domestic shipbuilding capacity. When the United States found itself at war and scrambling for ships to transport U.S. forces to France, its shipyard base could not implement a building program of the required magnitude.

New yards had to be built from scratch. The Hog Island shipyard across the Delaware River from Philadelphia rose from a mud flat in a matter of months to become the world’s largest shipyard, capable of simultaneously building fifty ships. To augment private yards, the Emergency Fleet Corporation opened four of its own: at Bristol, Rhode Island; Newark, New Jersey; Wilmington, North Carolina; and Hog Island itself — which collectively accounted for 25 percent of U.S. shipbuilding.

The desperation to secure ships manifested in other ways, too. The U.S. government, via the Emergency Fleet Corporation (EFC), awarded contracts to shipyards in China (the Kiangnan Dock & Engineering Works in Shanghai) and Japan for 45 ships totaling 375,000 tons. Furthermore, nearly a third of the EFC's 3,282-ship program — 1,017 vessels — was built from wood.

This is perhaps less surprising than it seems. Of 61 shipyards at the time the United States entered the war, 24 were devoted to wooden construction. The results were poor. Of an original program of 1,017 wooden vessels, shipyards had delivered just 87 by the Armistice — none of which had crossed the Atlantic (though 167 wood and composite vessels would make the voyage in 1919).

The shipbuilding problems were not limited to wooden vessels. According to the U.S. Shipping Board's own annual reports, although hundreds of steel ships would be built, the vast majority were delivered after hostilities had ceased. Of 1,693 steel ships delivered by the time the wartime shipbuilding program terminated, only 399 (roughly 24 percent) had been completed before November 1, 1918. Of those 399, fewer than one-third — 127 — were built from scratch, with the remainder consisting of ships already at various stages of completion that were requisitioned for the war effort. Not a single ship constructed at Hog Island was delivered in time to serve during the war.

The vast sums spent on new vessels — approximately $3.3 billion, or $60–65 billion in 2024 dollars — did not set the stage for a new era of competitive American shipbuilding. Indeed, the end of the wartime effort saw commercial output collapse and remain nearly dormant for most of the 1930s. In 1935 hearings before the House Committee on Merchant Marine and Fisheries, the Solicitor for the Department of Commerce, South Trimble Jr., testified that in the previous five years, U.S. shipyards had built just 4 freighters of 2,000 gross tons and over, and just another four in the five years before that. The equivalent figures for Great Britain were 295 and 558.

Further testimony drove the point home. H.G. Smith, president of the National Council of American Shipbuilders, told the committee that "in the United States there has been practically no cargo-vessel construction during the 10-year period, while all other important maritime nations have extensively built vessels of this type." During the same period, Britain renewed 41 percent of its cargo fleet. The figures, Smith concluded, were simply "deplorable." A report from the U.S. Chamber of Commerce's special committee on the merchant marine, included as an appendix to the hearing transcript, added that the industry had been "several times, during the past 10 years, on the verge of virtual disintegration due to lack of work."

By the time the Maritime Commission issued its 1938 annual report to Congress, matters had not improved. The U.S. shipbuilding industry was so depleted that the report noted that no ocean-going merchant vessels had been built on the Gulf or Pacific coasts since World War I.

Fassett's The Shipbuilding Business records the industry reaching its nadir in the mid-1930s with just four ships over 2,000 gross tons — two cargo ships and two tankers — delivered across 1934 and 1935 (p. 78).

The Maritime Commission's own 1939 Economic Survey of Coastwise and Intercoastal Shipping offered a matching verdict: with few exceptions, there had been "practically no vessels constructed for the coastwise and intercoastal trades subsequent to the completion of the Government's war-built fleet."

There were, of course, extenuating circumstances. The Great Depression no doubt contributed to the decline, as did the enormous amount of war-built tonnage sold off into the market at cut-rate prices. Between January 1, 1922 and August 30, 1934, 71 vessels built at a cost of approximately $200 per ton were sold at an average of just $20.98 per ton, a ten-to-one write-down. But these are far from the whole story. The failure of U.S. coastwise restrictions to foster robust levels of commercial ship construction was already a long-established theme.

Writing in The Atlantic in 1909 — twenty years before the Great Depression began — Winthrop Marvin observed that the great American shipyards "cannot be successfully maintained either by the demands of the present naval programme or by the relatively light and fitful work of the coastwise trade." [emphasis added] Fassett, meanwhile, noted that prior to 1914, 75 percent of U.S. shipbuilding capacity was found in naval work, not commercial construction. It was a dependency that industry insiders were still acknowledging a decade later.

In 1924 testimony before Congress, Captain Charles McAllister, Vice President of the American Bureau of Shipping, stated plainly that naval construction had "kept the yards fairly well employed" and that "the Navy has furnished the backbone of the shipyards." Merchant marine construction, he added, was "incidental, more or less."

This pattern persists: a 2021 MARAD report found that military shipbuilding and repairs accounted for 78.7 percent of U.S. shipbuilding and repair industry revenue in 2019. The heavy reliance on military contracts was further confirmed in 2024 by Sam Norton, CEO of Jones Act tanker company Overseas Shipholding Group, who noted that "the occasional vessel built to operate within the Jones Act is, for most [U.S. shipyards], a minor sideline interest."

Despite the dearth of commercial construction for most of the interwar period, two developments meant the U.S. shipbuilding industry was not starting entirely from scratch on the eve of World War II. The Merchant Marine Act of 1936 had established the Construction Differential Subsidy and created the Maritime Commission, which began placing orders for new merchant vessels and slowly reactivating yards that had fallen idle.

And in December 1940 — a full year before Pearl Harbor — Britain contracted with Todd Shipyards for 60 Ocean-class cargo ships worth $100 million, to be built in entirely new yards constructed for the purpose. The design, adapted from a British tramp steamer, became the direct template for the Liberty ship, and the yards built for the Ocean-class order went immediately into Liberty hull production once the U.S. entered the war.

The resulting effort was undeniably impressive, with American yards delivering roughly 5,000 cargo ships during the war, a feat of industrial mobilization without precedent. But that achievement did not obscure the fact that the United States had entered the war on the shipbuilding backfoot.

The Harvard University Graduate School of Business Administration, in a report prepared for the U.S. Navy Department and the U.S. Maritime Commission, was unequivocal: “…when the United States entered both World Wars the national security of the United States was seriously threatened because the country had failed to have a shipbuilding and a shipping industry suitable to its security requirements.” The American Merchant Marine Institute, meanwhile, acknowledged that “this country found itself at the outset of the conflict with a woefully inadequate fleet of American-flag ships and a lack of sufficient shipyard capability to build the necessary vessels.”

Two conclusions follow from the American shipbuilding experience in the world wars. First, requiring that vessels engaged in coastwise commerce be domestically built could not generate the volume or competitive pressure needed to sustain an industry capable of building ocean vessels. Second, those tempted to blame the Great Depression for shipbuilding’s interwar torpor must engage in some reflection. If the Jones Act’s build requirement can only be relied upon to produce shipbuilding when economic conditions are favorable, then it is not a durable national security asset but a fair-weather arrangement.

The Jones Act was not the only tool the United States employed to encourage its shipbuilding industry. The postwar period offered one more test, in the form of direct subsidies.

The postwar period offered one more test of whether subsidies could accomplish what protection alone had not. The Construction Differential Subsidy, formalized under the Merchant Marine Act of 1936 and expanded significantly by the Merchant Marine Act of 1970, was the most sustained and generous peacetime attempt to close the cost gap through direct federal intervention. At its peak the program paid the difference between American and foreign construction costs — up to a statutory ceiling of 50 percent of the vessel's price — for ships operating in international trade. Between 1936 and 1981 it disbursed $3.8 billion to American shipyards.

Although the program did prompt additional shipbuilding, simply covering the difference between U.S. and foreign prices proved an ineffective means of making U.S. shipbuilding competitive. By the late 1970s, with the program still running at full force, the cost gap had outgrown the subsidy’s legal reach. Foreign yards, particularly those in Japan, were quoting prices 60 percent below American levels. Avondale, widely considered the most efficient commercial yard in the country (since closed), required a 49.98 percent subsidy — essentially the legal ceiling — just to win a single contract in 1979. Overseas, two comparable ships could be built for the price of one American vessel.

The subsidies were discontinued in 1981, but even by then the commercial order book had already begun to crumble. MARAD's 1980 annual report disclosed that no CDS contracts had been awarded for new ship construction that year. The Secretary of Commerce had warned the year before that the industry's problems were of a magnitude that could not be overlooked. He wasn't alone in his warnings about the industry's outlook.

Contemporary press coverage documented the deterioration in real time. By 1978, Business Week was warning of imminent shipyard closures despite the subsidy program still running at full force. A Baltimore Sun investigation the following year found the industry in its most serious postwar crisis, with major yards expected to close one by one. By mid-1980, the New York Times reported only three commercial ship orders that year, and industry leaders were calling for government intervention to stave off collapse. The subsidies had not arrested the sector's decline but had merely slowed it.

The 1981 defunding of the CDS program accelerated the collapse of commercial shipbuilding, but the trajectory was already clear. A subsidy program hitting its legal ceiling while the cost gap continued to widen was not sustainable in the long term, and notions that U.S. commercial shipbuilding would be in a dramatically different place today with CDS in place should be greeted with skepticism.

Looking back across the sweep of U.S. commercial shipbuilding history, the Jones Act's record in promoting its fortunes is difficult to defend. The United States is the world's second-largest manufacturing economy and is renowned for its dynamism and innovation, yet it ranks just nineteenth in shipbuilding. It trails not just Asian giants but European countries whose industries operate without a protected domestic market. With commercial output already near zero, it strains credulity to argue a different policy path could have produced worse results.

The causes of decline are multifactorial, and it is very possible — even likely — that the United States would not have become a major shipbuilding nation under any policy regime. But the current record would not be difficult to improve upon. One can imagine American yards, freed from the artificial shelter of the Jones Act and the comfort of captive military contracts, having developed specialized niches and competing selectively in world markets. But that path was never taken.

The Jones Act's origins lie in an era when merchant vessels could be converted into privateers to hunt enemy shipping, and American commercial shipbuilding was competitive enough that protectionist shipping measures imposed relatively little cost. That world is long gone, and a law rooted in it is ill-suited for the twenty-first century.

What this project is. This site assembles primary-source data on U.S. commercial shipbuilding output, fleet size, vessel costs, and the cost premium of U.S.-built ships relative to foreign equivalents, and traces the history of American shipbuilding from the colonial era through the collapse of the Construction Differential Subsidy program in 1981. It draws analytical conclusions from that data and history, including about the Jones Act’s role in shaping — and failing to sustain — the U.S. shipbuilding industry. Every figure is sourced to a specific document; where estimates are used, the basis for the estimate is stated.

What this project is not. It is not a comprehensive economic model of the Jones Act’s costs and benefits. It does not attempt to quantify the full downstream effects of the shipbuilding cost premium on freight rates, consumer prices, or specific markets. It does not attempt a full accounting of national security tradeoffs, which involve considerations beyond the scope of shipbuilding cost and output data. It is a data-driven historical analysis, not a policy brief.

Conservative estimation throughout. Where a range of figures is available, this project uses the more conservative estimate. Where a source provides a lower-bound figure (e.g., Kinder Morgan's ">$240M"), that floor is used rather than a midpoint or upper estimate. This is consistent with the approach taken in the underlying research: claims are supported only to the degree the evidence warrants, and no figure is presented with more precision than the source justifies.

Primary sources over secondary. The default sourcing hierarchy is: (1) government data — MARAD, UNCTAD, CRS, GAO; (2) operator disclosures — annual reports, investor presentations, SEC filings; (3) industry data — BRS Group, classification society records; (4) contemporaneous press and congressional testimony. Secondary aggregators are used only where primary sources are unavailable or where they synthesize data across multiple primary sources.

The cost premium series (Section 03). The historical cost premium table spans 1775 to 2026 using heterogeneous sources: contemporary price data, government surveys, congressional testimony, operator disclosures, and CRS analysis. These are not drawn from a single consistent methodology. The 1950s–1961 entry is based on two independent sources (CRS R45725 and Senate testimony) that reach the same figure; earlier entries reflect different benchmarks (iron ships vs. British equivalents, contract prices, survey data). The table documents that a premium has existed continuously and grown substantially — it does not claim the figures are directly comparable across periods.

Fleet counts. The Jones Act eligible fleet figures use MARAD's vessel inventory reports as the primary source. "Eligible" means vessels that meet the four Jones Act requirements (U.S.-built, U.S.-documented, U.S.-owned, U.S.-crewed) and are in active oceangoing domestic trade. Vessels in layup, reserve, or government service are excluded. MARAD's reporting methodology has changed over time; figures before and after 2014 are not drawn from identical datasets. Fleet DWT comparison: 1951 fleet DWT of 5,333,000 is from the MARAD Annual Report series (see fleet DWT chart data). Current Jones Act fleet DWT of 4,819,648 is calculated from MARAD, United States Flag Privately-Owned Merchant Fleet Report, January 5, 2026, summing deadweight tonnage for all 93 Jones Act eligible vessels. Real GDP growth (more than nine times) is based on BEA GDPCA chained 2017 dollars via FRED, comparing 1951 to 2025.

Global output comparisons. World shipbuilding output figures use UNCTAD's Review of Maritime Transport and MARAD Annual Reports for the 1951–2001 period, and UNCTAD data for 2001–present. These measure gross tonnage or deadweight tonnage delivered; the U.S. share figures are calculated from those series. The series is not continuous across all years — missing data points are noted in the chart.

The China supply chain section (Section 05). Component origin data for individual vessels is drawn from ABS classification records, MARAD vessel construction documentation, USCG NVDC determination letters, and customs import records. These sources do not cover all components in all vessels; the data reflects what is documented, not necessarily a complete supply chain audit. The Caleb Petitt dataset covers 93 vessels tracking four component categories; it does not claim to be exhaustive across all component types.

Rounding and approximation. Figures presented as approximate (prefixed with ~) reflect either rounding of a precise figure or a range in the underlying source. Figures presented without approximation markers reflect the exact figure as stated in the primary source. Per-TEU and per-DWT cost calculations are derived from reported vessel costs and design capacity; actual unit economics will vary with utilization.

Last updated. Data current as of June 2026. Sources with specific publication dates are noted in chart source lines. MARAD fleet data reflects the January 2026 vessel inventory report.