A 70-Year History of Container Shipping

01The Confusion Starts with How We Count Boxes—The Trap of TEU and Handling Costs

Containers basically come in two sizes: 20-foot and 40-foot. Here, counting one 20-foot container as 1 gives us the unit known as TEU. A 40-foot container is counted as 2 TEU. When the news mentions a "20,000 TEU-class ship," it means a vessel with space for the equivalent of 20,000 twenty-foot boxes.

There are two things to watch out for here. First, TEU is a unit of space, not weight. Pack it with feathers, and it is 1 TEU; pack it with bolts, and it is still 1 TEU. That is why a ship's rated TEU capacity is only a theoretical maximum. If the cargo is heavy, draft limits—how deep the ship sits in the water—and balance constraints hit first, keeping the vessel from filling every slot. Writing that a "24,000 TEU ship" always carries 24,000 containers is simply incorrect.

Second, port throughput statistics count containers multiple times. When giant ships call only at major hub ports and transfer cargo onto smaller vessels bound for nearby regional ports—a process known as transshipment—a single box is counted when unloaded and counted again when reloaded. This transshipment volume is a major reason why Singapore and Busan rank so high in global throughput. So hearing that "world ports handled 920 million TEU" actually reflects a figure that counts the same box multiple times, not that there are 900 million containers in the world.

The scale itself is staggering. In 2024, global seaborne cargo volume totaled about 12.3 billion tons, combining crude oil, grain, and ores. By weight, container cargo is only a fraction of that. But when you look at value, the story completely changes. Cargo carried by member companies of the World Shipping Council alone accounts for just over 60% of the value of global maritime trade, exceeding $4 trillion a year. A minority by weight, but the vast majority by value—that is what fills these containers.

And throughout this story, there is one distinction you must keep in mind: the cargo handling costs for loading and unloading ships, the freight rate paid by shippers to shipping lines, and the total logistics costs, which include inland transport, warehousing, and inventory. These three are entirely different, and the answer to what the container changed—and by how much—differs for each one.

Worth remembering The easiest place to get tripped up in this story isn't the years, but the units. Separate TEUs, handling costs, and freight rates first, and everything else stays clear.

02Hooks, Coffee Sacks, and the Days Ships Were Tied Up in Port

Before containers, general cargo shipping was known as breakbulk. While bulk cargo refers to loose goods scooped up like grain or coal, breakbulk is cargo counted piece by piece. Sacks of coffee, bales of cotton, steel drums, and wooden crates. Dockworkers stacked them one by one inside the ship using hooks and cargo nets.

This was not simple labor. Because every piece of cargo varied in weight and shape, workers had to judge how to stow them so the ship would not list, deciding what went on the bottom and what stayed on top. The seasoned eye of an experienced worker was itself the core skill. But it was terribly slow. A single work gang reportedly moved only about 5 to 10 tons an hour. Today, a modern gantry crane handles around 500 tons in that same amount of time.

Slow meant expensive. According to one estimate, about half of the total freight cost was spent loading and unloading cargo at the port, while the actual voyage across the ocean accounted for only around 11% of the total cost. The original source does not specify which cargo or route this figure was calculated for, but the overall picture is clear: the real expense was accumulating at the pier, not at sea.

Cargo theft and damage were also routine. Goods vanished or broke every time shipments were shifted and handled. That was why port cities were built around their piers. Along the waterfronts of Manhattan and Brooklyn, finger piers jutted out side by side in tight rows, with warehouses, saloons, and worker tenements stretching right behind them. Every morning, crowds of men gathered in front of the docks hoping to land a day's work.

Yet the idea of 'putting cargo in boxes to move it' was remarkably old. In 1766, English canal engineer James Brindley designed a boat carrying ten wooden boxes to haul coal. In 1928, the Pennsylvania Railroad launched a container service connecting 8 major US cities, and in 1932, it even opened the very first container terminal. The US military developed the Transporter in 1947 and the CONEX system in 1952, shipping cargo all the way from Georgia to Korea.

The boxes were already there. But those boxes never traveled beyond their own fences. A railroad company's boxes circulated only within that railroad, and military boxes moved only within the military. Non-standardized dimensions, railway freight rate regulations, and entrenched dock labor practices are commonly cited as the tangled reasons, though no study has definitively settled which factor was decisive. A box's true value comes not from how sturdy it is, but from its transferability across different modes of transit—and as of yet, no one had a reason to push that transferability beyond their own backyard.

Worth remembering Before the container, the issue was not the technology—it was where the costs were occurring. Knowing this makes it clear why the math in the next chapter worked.

03Why Did a Trucking Boss Sell His Trucks and Buy Ships?

In 1935, Malcolm McLean, a 21-year-old pumping gas at a North Carolina service station, started a trucking company with his siblings using a single secondhand truck. In 20 years, McLean Trucking grew into one of the leading carriers in the United States. In other words, he was someone who knew very little about ships. And that turned out to be decisive.

Around 1953 to 1954, McLean looked into loading whole trucks onto ships to bypass congestion on US East Coast highways. Then he changed his mind. Loading wheels and chassis meant loading empty air inside the ship, so why not detach the box and load only that? While dramatic anecdotes circulate in several versions about how the idea struck him while waiting for cargo to be unloaded at a dock, there is no confirmed testimony. What is certain is that he was calculating not the cost of the ship, but the door-to-door cost.

Yet becoming a shipowner was not entirely a free choice. In 1955, the Interstate Commerce Commission prohibited a single person from controlling both trucking and shipping operations at the same time. He had to choose one or the other. McLean sold his 75% stake in McLean Trucking for $6 million and acquired Pan-Atlantic Steamship Corporation. In May of that year, the Interstate Commerce Commission granted Pan-Atlantic a 180-day temporary operating permit for maritime transport. Both the narrative that regulation stifled innovation and the narrative that regulation pushed an innovator into an unexpected place stem from this very same fact.

On April 26, 1956, the Ideal-X, a converted World War II tanker, departed Port Newark, New Jersey. It carried 58 containers on its deck alongside 15,000 tons of oil. Five days later, it arrived in Houston, Texas, and unloading took less than 8 hours. Cargo orders for the return trip arrived before the vessel was even fully unloaded. For reference, sources differ on the length of these boxes, with some citing 33 feet and others 35 feet.

A famous set of numbers is attached to this voyage: the calculation that loading and unloading a conventional breakbulk freighter cost $5.83 per ton, whereas on the Ideal-X it was 15.8 cents. While widely cited, it must be handled with caution. This was not the freight rate, but the stevedoring cost of loading and unloading a ship, calculated by McLean's team for a single voyage of a single vessel. Furthermore, nearly everyone is simply repeating the figures published in Marc Levinson's book; no verifiable primary sources have been released, and the figures vary slightly across documents.

In October 1957, the following year, the Gateway City left Newark for Miami. Featuring vertical steel cell guides fitted inside the hold to stack boxes up to five tiers high, it is recognized as the world's first cellular containership. It carried 226 containers, and the cranes mounted on the ship could lift 60,000 pounds. The industry had moved from placing boxes on a deck to designing the ship itself around the box.

Worth remembering Innovation came from outside shipping, not within it—and it was regulation that pushed him from the outside in.

04The Real Invention Wasn't the Box, but the Steel at the Corners

In 1955, McLean brought in a mechanical engineer named Keith Tantlinger. What Tantlinger designed was not the box itself, but its corners. He attached hollow steel blocks with holes to all eight corners (corner castings) and created a mechanism that inserted into the hole and locked with a 90-degree turn (twistlocks). A crane lifting a box, or ten boxes stacked and secured aboard a rolling ship—everything relied on this single piece of hardware. Container standardization was, in essence, the world agreeing on the exact positions and dimensions of these corner steel pieces.

Around the same time, another company entered the business through an entirely different approach. Matson, a shipping line serving Hawaiian routes, established an in-house research team to quantitatively analyze trade routes, cargo flows, and optimal box sizes. In August 1958, the Hawaiian Merchant departed San Francisco carrying 20 24-foot containers. The following year, the company erected the world's first A-frame gantry crane in Alameda, California. Standing on the pier rather than aboard the ship to lift boxes, this crane became the prototype for every modern container terminal. The popular narrative that a single genius's intuition changed the world begins to waver right here.

The real issue was that every company used a different box. Sea-Land's boxes could only be hoisted by Sea-Land's cranes, and Matson's containers engaged only with Matson's gear. If cargo had to be unstowed and repacked every time it shifted ports or carriers, the entire purpose of containerization was defeated.

In 1961, the International Organization for Standardization formed Technical Committee TC104 to establish container standards. Over the next several years, the United States, pushing for 35-foot and 40-foot lengths, clashed with European nations that preferred smaller boxes. While verbatim transcripts of those meetings do not survive, the lines of battle were clear. Meanwhile, American shipping lines were already forging across oceans. In April 1966, Sea-Land's Fairland sailed from Elizabeth carrying 226 35-foot containers, arriving in Rotterdam 10 days later. While widely remembered as the first transatlantic container voyage, another carrier inaugurated a scheduled service to Europe in that exact same month, meaning what counts as the "first" depends entirely on definition.

Agreement ultimately arrived in writing. In January 1968, ISO 668 was published, defining the terminology and external dimensions of Series 1 containers. In July of that same year came the identification marking standard, followed by standards for corner fittings in January 1970 and minimum internal dimensions in October 1970. That a single box could now be locked into identical equipment anywhere in the world was thanks to these three or four documents.

There is a common misunderstanding here: the myth that "McLean made all his patents freely available to the public." The verified record is far narrower. Tantlinger persuaded McLean to permit the royalty-free use of Sea-Land patents covering corner fittings and twistlocks strictly to achieve standardization. No specific surviving legal documents spell out the exact patent numbers or licensing terms. Yet the core truth remains: If the standard had remained the proprietary property of a single company, the container could never have become the world's universal language.

The final link in the chain was forged on land. In July 1977, the first double-stack railcar was delivered in the United States, and by 1981, five-unit well cars became the industry standard. By stacking boxes two high straight off the ship to cross the continent, the seamless transport chain—transferring freight from truck to ship and ship to train without ever rehandling the cargo inside—was at last complete.

Worth remembering This is the turning point where the hero of the story shifts from the individual inventor to the universal standard. Until the boxes were made identical, a container was just a sturdy crate.

05The Big Player That Spread Containers Worldwide Wasn’t Merchants—It Was the Military

The transatlantic routes had begun to open, but well into the mid-1960s, the transpacific side remained largely empty. To fill boxes on an ocean route, you need cargo on both ends. Yet to justify the massive investment needed for cranes and container yards at new ports, that cargo volume had to be guaranteed first. It was a classic chicken-and-egg problem.

The one who untied this knot wasn't commercial traders, but the U.S. military. The number of CONEX boxes held by the U.S. military surged from 100,000 in 1965 to over 200,000 in 1967. And in 1967, Sea-Land signed a contract to manage full container logistics for the U.S. military supply base in Cam Ranh Bay, Vietnam, even installing container cranes on site. They also operated on a smaller scale in Da Nang.

With the government guaranteeing cargo volume, risky capital investments became viable. A military contract guaranteeing cargo volume absorbed the steep initial costs that commercial shippers alone could never have justified. But the real consequence unfolded right after. Instead of returning to the United States empty after unloading military supplies, the ships stopped by Japan to take on commercial cargo. This account—that transpacific container liner services were born this way—is widely cited through Levinson’s work. However, concrete figures showing the exact volume of that return cargo remain unverified.

Around the same time, the board was shifting on the other side of the ocean as well. In June 1967, the Suez Canal was shut down and did not reopen until 1975. During those 8 years, ships sailing between Asia and Europe had to route around Africa. It is hard to say definitively that this closure accelerated containerization. Still, as voyages grew longer, carrying more in a single trip became vastly more advantageous, establishing the logic of building bigger ships to endure long hauls as an accepted norm across the industry.

This is also where South Korea enters the story. In 1978, the Jasungdae Container Terminal opened at the Port of Busan. It was Korea’s first dedicated container terminal, serving as the physical conduit for a country aiming to sell manufactured goods to the world. No matter how many factories you build, export manufacturing cannot survive without docks capable of shipping cargo out on a reliable schedule.

That is why reading the story of the container merely as a "Western tale that started in the U.S. and Europe and spread to Asia" misses half the picture. What opened the transpacific route was the supply demand of the U.S. military stationed in Asia, and what filled that route were Asia’s factories. Today, ports in developing Asian nations handle 60% of global container throughput. In 2025 alone, the Port of Shanghai handled 55.06 million TEU, ranking first in the world for 16 consecutive years.

Worth remembering The trigger for global adoption was government cargo, not the open market—and its byproduct laid the groundwork for Asia’s export boom.

06What Took the Place of the Vanished Wharves?

A container port looks entirely different from a breakbulk port. For a single crane to unload boxes without pause, there needs to be a vast marshalling yard right behind the wharf to stack them. But the old finger piers jutting out along downtown riverfronts simply didn't have that kind of land. And as ships grew larger, water depth became an issue too.

So the ports packed up and moved. In August 1962, the Port Authority of New York and New Jersey opened the world's first purpose-built container facility in Elizabeth, New Jersey. This layout, anchored by a vast yard behind the docks, became the standard for every container port that followed. Meanwhile, the wharves of Manhattan and Brooklyn lost their business. It was a move of just a few kilometers across the river, yet an entire city's industry was handed over.

Labor had already arrived at the bargaining table before that. In October 1960, the West Coast dockworkers' union, the ILWU, led by Harry Bridges, signed the Mechanization and Modernization Agreement with employers. The union surrendered long-standing work rules and agreed not to block machines; in return, workers received a 35-hour workweek and a $29 million early-retirement pension fund. Whether to view this as a surrender or a shrewd deal was fiercely debated within the union even back then. After all, conditions improved for existing members, but future job openings for the next generation shrank.

Something even more dramatic took place in London. Beginning with the East India Dock in 1967, docks shut down one after another, until the last ship loaded cargo in December 1981. Blaming this entirely on the container, however, would be an exaggeration. The old upstream docks were too shallow for the new ships, protracted labor strife drove cargo owners to other ports, and British manufacturing itself was in decline. All three causes converged at the exact same time.

The impact, in any case, was massive. Although a 1972 agreement in the UK protected dockworkers from compulsory redundancy, the number of registered dockworkers plummeted over the next 20 years. A drop from 60,000 to under 10,000 is widely cited, though the original statistics are hard to verify. When the Thatcher government abolished the National Dock Labour Scheme in 1989, that protective barrier vanished as well. The Canary Wharf financial district that rose on the site of the London Docks is simply another name for that empty space.

This story isn't just a thing of the past. In October 2024, the East and Gulf Coast dockworkers' union, the ILA, staged a three-day strike. The sticking point wasn't wages, but semi-automated cranes. The tentative agreement reached in January 2025 blocked full automation while permitting semi-automation, provided the union was guaranteed the jobs tied to the new technology. The structure was almost identical to the deal struck on the West Coast over 60 years earlier. Today's debates over automation are nothing new.

Worth remembering The bill for the container was footed not by cargo owners, but by port cities—and by the next generation of workers who never made it through the gates.

07The Gamble of Building Bigger Ships, and the Bankruptcy of the Man Who Championed the Container

Once the boxes became identical, the next competition shifted to size. As a ship gets larger, the fuel and labor costs per box decrease. Few industries demonstrate economies of scale quite so nakedly. The Triple-E class launched by Maersk in 2013 had a nominal capacity of 18,270 TEU and a design speed of 23 knots, and calculations indicated that its fuel cost per TEU was about 35% lower compared to the earlier 13,100 TEU class. Of course, this figure assumed specific fuel prices and routes.

Yet this calculation hides a trap. Building a mega-ship takes years, and whether to build it or not must be decided based on today's fuel prices and cargo volumes. There is no guarantee that the world at the time the ship arrives will be the same as when it was ordered.

The most famous victim of this gamble was the very man who had championed container shipping. In 1969, McLean sold Sea-Land to the tobacco and food giant R.J. Reynolds for $530 million. By then, Sea-Land was operating 27,000 containers and calling at more than 30 ports. He reentered the game in 1979 by acquiring US Lines and ordered twelve 4,000 TEU "Econships" from Daewoo Shipbuilding, designed with reduced speeds to save fuel. The premise was that oil prices would keep rising. By the time they were delivered, fuel prices had plunged, leaving them as merely slow ships. In 1986, US Lines went bankrupt. McLean passed away in New York in 2001, and story has it that on the morning of his funeral, container ships in ports around the world sounded their whistles.

For a long time, ship sizes were dictated by the locks of the Panama Canal. The maximum size that could pass through the old locks was called Panamax, usually about 4,000 TEU, stretching to around 4,500 TEU when a ship was designed to fit the locks exactly. The race began in earnest in 2006 when Maersk's Emma Maersk became the world's largest container ship at 397 meters in length. Capacity figures vary widely depending on the source, but it is generally known as roughly the 15,000 TEU class. In June 2016, the expanded Panama Canal opened. Costing a total project budget of $5.25 billion and finishing nearly two years behind schedule, the new locks were 21 meters wider and over 5 meters deeper than the old ones, allowing 366-meter ships to pass through. The first vessel to transit was a ship from China's COSCO.

South Korea entered this race as well. Delivered in April 2020, the HMM Algeciras had a nominal capacity of 23,964 TEU, a length of 399.9 meters, and a beam of 61 meters, making it the world's largest at the time of delivery. As of mid-2026, the MSC Irina class is cited as the current largest at about 24,346 TEU, and the top 20 ships all exceed 24,000 TEU. It is a number that must be described not as "the largest ever," but as "the largest right now." What is intriguing is the ceiling. Even as ships kept growing, they have yet to breach the 25,000 TEU mark.

Growing bigger carries another implication: it means giving up on a certain canal. The very largest vessels are permanently assigned to the routes linking Asia and Europe. And that route is held by the throat at narrow waterways.

Worth remembering Economies of scale are both the engine and the gamble of this industry, and the most dramatic casualty of that gamble was the very man who championed the container.

08The Bill That Arrives in Narrow Waters

On March 23, 2021, the 400-meter-long Ever Given wedged itself across the Suez Canal during a sandstorm. By the time it was refloated six days later, more than 400 ships were backed up waiting. At the time, the phrase "$9 billion a day in losses" spread widely, but that was not the actual loss—it was an estimate of the value of cargo passing through the waterway each day. Most of that cargo was never lost; it was simply delayed. We need to distinguish between the value of a transit corridor and the damages of an accident.

The following year brought a different kind of bottleneck. On January 9, 2022, an all-time record of 109 container ships were waiting to unload off the coast of Los Angeles and Long Beach. That same month, the Shanghai Containerized Freight Index hit an all-time high of 5,109.60 points. It was the combined result of pandemic-driven consumer spending surges and severe shortages of dockworkers and equipment. Note that the count of waiting ships varies depending on the tracking agency and how waiting zones are defined, and this index reflects only spot freight rates out of Shanghai, which differ from long-term contract rates or total logistics costs.

In 2023, the skies sent a bill of their own. A severe drought caused water levels in Lake Gatun in the Panama Canal to drop to their lowest since 1965. Normal daily transits of 36 to 38 ships shrank to 18 to 24 ships a day by early 2024, depending on the data source. Normal capacity was not restored until August 2024. Because the canal's lock system flushes lake water into the sea each time a vessel passes, there is simply no choice but to cut traffic when the rains do not come.

Around the very same time, the Red Sea was choked off. Starting in November 2023, the Houthis in Yemen began attacking merchant vessels, forcing most Asia-Europe container ships to detour around the Cape of Good Hope, which lengthened voyages by 30 to 50%. Daily cargo volume moving through Suez fell from 4 million tons a day in early 2024 to around 1.7 million tons. As of the summer of 2026, the route is cautiously reopening. Maersk announced that more than 30% of its diverted Asia-Europe volume has returned to Suez, and other carriers have restored some services, but traffic remains about 40% below pre-crisis levels.

Waterways are not the only point of vulnerability. On August 31, 2016, Hanjin Shipping filed for court receivership. It was the world's 7th-largest ocean carrier, controlling 3.2% of global container capacity. Fearing they would not be paid for unloading cargo, ports turned ships away, stranding 96 vessels at sea. The fate of about $14 billion worth of cargo and 540,000 containers remained in limbo for weeks. For an export-dependent nation, it was a painful lesson: if a ship cannot dock at port, the goods inside it come to a dead halt as well.

Today, the industry is concentrated in just a few hands. As of mid-2026, MSC alone commands 7.34 million TEU of capacity, holding a 21.6% global market share, with a gap of more than 2.6 million TEU over second-place Maersk. In February 2025, the alliance map was redrawn as well. Maersk and Hapag-Lloyd launched the Gemini Cooperation, running services with 290 ships; HMM, ONE, and Yang Ming formed the Premier Alliance; and MSC chose to sail alone. Pushing efficiency to its absolute limit eliminates all slack, and when there is no slack, a single incident quickly turns into a worldwide delay.

Worth remembering The supply chain crises we see on the news are not accidents, but the result of design. The bigger we build the system, the steeper the bill that comes due in narrow bottlenecks.

09So, Did Shipping Costs Really Go Down?

Now comes the hardest question of this story: Did containers really change trade?

The most frequently cited study is a 2016 paper by Bernhofen, El-Sahli, and Kneller. Using data from 1962 to 1990, they estimated that trade between two countries increased by 320% in the year both had container ports. However, we should not read this figure as "containers increased world trade by 320%." It is an estimate comparing pairs of countries that adopted container facilities with those that did not, and the authors themselves discuss the possibility that other factors expanding trade may have been captured at the same time.

On the other side stands David Hummels. In a 2007 paper, he estimated that when the containerized share of trade doubles, shipping costs decrease by 3 to 13%. What is even more striking is what comes next: he stated that he found no clear evidence of a decline in liner freight rate indices over the past 50 years. His explanation is that rising fuel costs, vessel prices, and port fees offset productivity gains.

Benjamin Bridgman takes this a step further. While containerization clearly triggered an explosion in port labor productivity, he asks why those gains did not trickle down to shippers. He argues that the bargaining power of strong longshore unions and the market power of liner cartels captured that share. In a sense, the M&M Agreement and the alliances we saw in previous chapters reappear right here.

When you place the magnitude of these numbers side by side, a crack appears. Estimates of trade growth run in the hundreds of percent, while estimates of shipping cost reductions are, at most, in the low teens. They differ by entire orders of magnitude. So it is more accurate to sum it up this way: what the container unquestionably broke down was cargo-handling costs. Whether freight rates fell by just as much remains under debate. The true savings in total logistics costs were far more likely to come from time than price. If you can predict when a ship will arrive, you do not have to stockpile as much inventory in warehouses. What the box changed was not the price tag, but the reliability of the promise.

What should we watch going forward? First are the narrow waterways. Whether transit through Suez returns to pre-crisis levels and how Panama weathers the next drought will show up immediately in freight rates. Second are automation negotiations. Opened in December 2017, Phase 4 of Shanghai's Yangshan Port handles 6.3 million TEU a year with remote-controlled cranes and automated guided vehicles. While Rotterdam was the world's first automated terminal in 1993, the center of gravity has shifted to Asia. How port labor prices this shift has been the same question for 60 years. Third is fuel. In September 2023, the world's first methanol-powered container ship was named in Copenhagen, and that same year the International Maritime Organization adopted a strategy to cut greenhouse gases by at least 20% by 2030, by 70% by 2040, and achieve net-zero around 2050. When fuel changes, ship sizes and route calculations will be completely redrawn.

What the 58 boxes that left Newark 70 years ago left behind is not a single sentence about how much cheaper things became. It was an experiment in what happens when different companies, different countries, and different modes of transport agree on the same standard. Look around your room right now. Almost everything that crossed the ocean got here through that very agreement.

Worth remembering Not a prophecy, but a spectator's guide: narrow waterways, automation negotiations, and the fuel transition—these three are the windows showing where the next cost structure will be decided.