Ocean carriers added 272 containerships to their fleets between 1976 and 1979. Four times during the 1970s, worldwide container shipping capacity increased by more than 20 percent in a single year. Total cargo capacity aboard containerships, 1.9 million tons in 1970, reached 10 million in 1980, not counting the tonnage of vessels designed for a mix of containers and other freight.4
The quest for scale brought not just more ships but bigger ships. The Fairland, the first Sea-Land ship to cross the Atlantic in 1966, was only 469 feet long. The purpose-built containerships of the late 1960s were about 600 feet from stem to stern, and the fast vessels launched in 1972–73 were as much as 900 feet long and 80 feet wide, with drafts of 40 feet. At that point, containership design seemed to be approaching its limits. The locks of the Panama Canal, through which almost all traffic between Asia and the Atlantic coast of North America had to travel, are 1,000 feet long and 110 across, and bigger ships would not fit. The oil crisis, which caused so many financial problems for ship lines, unexpectedly brought relief. Shipowners decided to build slower vessels to save fuel: the average speed of newly delivered containerships dropped steadily from 25 knots in 1973 to 20 in 1984. Naval architects were no longer forced to design streamlined shapes to help achieve high speeds, and could concentrate instead on increasing payloads. Without getting much longer, vessels got much larger. The ships entering service by 1978 could hold up to 3,500 20-foot containers—more than had entered all U.S. ports combined during an average week in 1968.
These Panamax vessels—the maximum size that could fit through the Panama Canal—could haul a container at much lower cost than could their predecessors. The construction cost itself was lower, relative to capacity: a vessel to carry 3,000 containers did not require twice as much steel or twice as large an engine as a vessel to carry 1,500. Given the extent of automation on board the new vessels, a larger ship did not require a larger crew, so crew wages per container were much lower. Fuel consumption did not increase proportionately with the vessel’s size. By the 1980s, new ships holding the equivalent of 4,200 20-foot containers could move a ton of cargo at 40 percent less than could a ship built for 3,000 containers and at one-third the cost of a vessel designed for 1,800.5
And still the vessels grew. The economies of scale were so clear, and so large, that in 1988 ship lines began buying vessels too wide to fit through the Panama Canal. These so-called Post-Panamax ships needed deeper water and longer piers than many ports could offer. They were uneconomic to run on most of the world’s shipping lanes. They offered no flexibility, but they could do one thing very well. On a busy route between two large, deep harbors, such as Hong Kong and Los Angeles or Singapore and Rotterdam, they could sail back and forth, with a brief stop at each end, moving freight more cheaply than any other vehicles ever built. By the start of the twenty-first century, ship lines were ordering vessels able to carry 10,000 20-foot containers, or 5,000 standard 40-footers, and even bigger ships were on the drawing boards.
As ships got bigger, ports got bigger. In 1970, the equivalent of 292,000 loaded 20-foot containers passed across the piers at Newark and Elizabeth, far and away the world’s largest container complex. In 1980, the wharves around New York Harbor, including the new U.S. Lines terminal on Staten Island, handled seven times that many loaded boxes, even though New York’s share of all U.S. container traffic had declined. Container traffic from Britain to points outside Europe, almost all of which passed through either Felixstowe or Tilbury, more than trebled in a decade, despite Britain’s weak economy. Deep-sea ports from Rotterdam, Antwerp, and Hamburg to Hong Kong, Yokohama, and Kaohsiung, on Taiwan, more than doubled the number of boxes they handled in the late 1970s. More and more, the biggest ports traded largely with one another: in 1976, nearly one-quarter of all U.S. containerized foreign trade went through Kobe, Japan, or Rotterdam, in the Netherlands, and another quarter went through just five Asian or European ports.6
The ceaseless expansion of port capacity was driven by the same force as the ceaseless increase in ship capacity, the demand for lower cost per box. New ships sold for as much as $60 million apiece in the late 1970s, despite the depression in shipbuilding. To cover their mortgage payments, ship lines had to maximize the time that their vessels were under way, filled with revenue-generating cargo, and minimize the time spent in port. The equation was simple: the bigger the port, the bigger the vessels it could handle and the faster it could empty them, reload them, and send them back out to sea. Bigger ports were likely to have deeper berths, more and faster cranes, better technology to keep track of all the boxes, and better road and rail services to move freight in and out. The more boxes a port was equipped to handle, the lower its cost per box was likely to be. As one study concluded bluntly, “Size matters.”7
Size mattered, but a port’s location mattered less and less. Traditionally, ports had prospered from interrupting the flow of trade. Customs brokerage, wholesaling, and distribution had been concentrated in port cities, as they were in New York, because all inbound and outbound cargo made a stop there. A port usually had overwhelming financial and commercial links with the interior region that was its hinterland. Geographers, once upon a time, had designated inland points as “tributary” to a particular port.
There were no tributaries in container shipping. Containers turned ports into mere “load centers,” places through which large amounts of cargo flowed with hardly a break. Each ship line organized its operations around a small number of load centers to minimize the number of stops its costly vessels would make. Customers did not care where those load centers happened to be: an Illinois manufacturer shipping machinery to Korea was indifferent as to whether its goods went by truck to Long Beach or by rail to Seattle, much less whether they entered Korea at Busan or Inchon. The ship line would make those decisions at its discretion—and it would make them based entirely on which combination of vessel operating costs, port charges, and ground transport rates would lead to the lowest total cost per box.8
This new maritime geography brought decidedly nontraditional trade patterns. Exports from southern France might move most cheaply through Le Havre, on the English Channel. Imports for Scotland might ride the train from southeastern England. Japanese cargo headed for San Francisco Bay might well be imported through Seattle rather than Oakland, with the ship line’s saving of one day’s steaming time in each direction outweighing the cost of putting some of the cargo on a train from Seattle to California. Port cities along the Gulf of Mexico increasingly did their European and Asian trade through Charleston or Los Angeles, because ship lines deemed sailing to the Gulf uneconomic. The Hampton Roads of Virginia displaced Baltimore as a major load center through no fault of Baltimore’s, but because a ship line serving Europe could make four more trips per year from Hampton Roads—and when a $60 million ship was involved, those four trips could spell the difference between profit and loss.9
The local economic benefits of a successful port still were large. A metropolitan area with a port would have a concentration of jobs in trucking, railroading, and warehousing, would need customs brokers and freight forwarders, and would reap tax revenue from port-related businesses. Where those jobs happened to be, however, depended upon commercial considerations much more than on geography. Ports such as Seattle, with small local markets, had a realistic hope of being major container gateways, “reaping substantial economic benefits as a result,” one study argued. Ports such as Tokyo and London, with populous markets close at hand, were not guaranteed to prosper. With the ship lines calling the tune, ports were forced to compete for the carriers’ favor.10
Competition involved investment on a dizzying scale. The World Bank and the Asian Development Bank poured $1.3 billion into port projects in developing countries during the 1970s. American ports spent $2.3 billion for container-handling facilities between 1973 and 1989. Ship lines used their bargaining power to push the risks of building new berths, buying new cranes, and digging deeper channels on to government-run po
rt agencies. The ports insisted that the ship lines sign leases, but leases often did not guarantee a flow of cargo. Ship lines could, and often did, move their load centers from one port to another as their strategies changed, making only minimal payments to the ports they left behind. In a single year, thirty ship lines shifted their service at North American ports, leaving some with a severe loss of traffic. Having the fanciest facilities did not guarantee success; Oakland spent a disproportionate share of its revenues for container terminals in the late 1970s but lost enormous market share to Long Beach nonetheless. In 1983, after yet more huge investments, an environmental lawsuit brought a dredging project to a halt, and American President Lines responded by moving much of its traffic to Seattle. Seattle, in turn, lost out in 1985, when Tacoma, a few miles south on Puget Sound, built a $44 million terminal and lured Sea-Land away.11
Inevitably, much of the investment in port facilities went to waste. Baltimore’s new docks led to a surge of cargo in 1979 and 1980—but the port handled fewer containers in 2000 than it had two decades earlier. Taiwan’s expensive containerport at Kaohsiung was a roaring success, but the government’s decision to build a port at Taichung as well was a costly mistake. San Diego was one of many ports to order high-priced container cranes that then found little use. Entire technologies, such as on-dock rail, proved to be sinkholes: ports that laid train tracks on the docks, so that cranes could transfer cargo directly from ships to waiting railcars, learned that the time required to move the train forward as the crane loaded each railcar delayed ships and reduced productivity. Many of the on-dock rail lines were abandoned, but the cost of failure was largely borne by the ports.12
The increasing riskiness of the port business did not pass unnoticed. Government investment in ports had been crucial to the development of container shipping in the 1960s and 1970s. With the exceptions of Felixstowe and Hong Kong, every major containerport in that era was developed at public risk and expense. At the time, there had been no alternative: the undercapitalized ship lines and stevedoring companies could not possibly have financed port development on their own. As investment needs grew larger, public officials began to lose their enthusiasm for running ports. “The incremental costs are now staggering,” the head of Seattle’s port said in 1981. The possibility that a ship line’s departure or demise could leave a public agency to pay for idle cranes and silent container yards was too great for many governments to chance.13
British prime minister Margaret Thatcher broke the ice by selling off twenty-one ports to a private company in 1981. Governments in other countries followed suit. Malaysia leased its container terminal at Port Klang to a private group in 1986, and ports from Mexico to Korea to New Zealand were soon in private hands. The investors included not only stevedoring and transport companies, but also leading ocean carriers. Containership lines were by now huge businesses, able to raise the large amounts of capital that ports required. As port users, they had an interest in having facilities that could handle their ships quickly. Unlike government agencies, the private port operators had no imperative to expand for the sake of local economic development; they could insist on long-term contracts, backed by banks or by collateral, to assure that they would recover whatever investments they made. Governments retreated to the role of landlords, renting out waterfront land to private companies. By the end of the twentieth century, nearly half the world’s trade in containers would be passing through privately controlled ports.14
In 1977, container shipping reached a landmark. Containerships were put into service between South Africa and Europe, the last big maritime route still handled by breakbulk ships. Containers were by no means universal; on many less trafficked routes, especially to Africa and Latin America, traditional ships still dominated. In commercial terms, though, these were niche markets, not large opportunities. The major ocean routes had become the floating highways that Malcom McLean had envisioned. Seventeen ships with a total capacity of 20,000 20-foot containers sailed each week from the U.S. Pacific coast to Japan in 1980. From northern Europe, 23 ships set sail weekly for Atlantic and Great Lakes ports in North America, and 8 more, with a capacity of more than 15,000 containers, left Europe for Japan. Even the long route between Australia and the U.S. East Coast had an average of 2.5 containerships in each direction each week, carrying Australian meat to America and manufactured goods the other way.15
In their endless quest to get bigger, ship lines set their sights on a new way of linking ports that they already served: sailing around the world.
Round-the-world service had been an idea hardly worth contemplating in the days of breakbulk shipping. With slow ships and long port calls, the 39,000-mile trip from New York across the North Atlantic, through the Straits of Gibraltar and the Suez Canal, calling at Singapore and Hong Kong and Yokohama and Los Angeles and heading home through the Panama Canal, would easily have taken six months. Faster vessels and shorter port calls made a three-month voyage imaginable. In 1978, the year after McLean’s departure, R. J. Reynolds ordered 12 diesel-powered vessels at a cost of $580 million and promised that Sea-Land would soon launch a “new weekly round-the-world service.”16
The idea was not entirely insane. Most ship lines suffered from highly imbalanced traffic patterns. Sea-Land, for example, was a major carrier in the North Pacific, but Japan’s huge trade surplus with the United States meant that it carried far more cargo east-bound than westbound. It had the reverse problem in the Middle East, where countries flush with petroleum revenues were importing vast quantities of manufactured goods but had little containerized freight to export. A service sailing eastbound around the world could help resolve this imbalance by allowing ships to discharge full containers at Middle Eastern ports, take on empties that had been delivered on previous voyages, and carry them onward to Japan. On the way, the vessels might stop in Singapore and Hong Kong, where they would be met by smaller ships shuttling freight from developing economies, such as India and Thailand, that did not yet trade enough to justify a container route to Japan or the United States.
Yet round-the-world service was a risky venture. Traffic flows between different pairs of ports were vastly different; a vessel that might be perfect for loads between New York and Rotterdam could easily be too large between Singapore and Hong Kong. Delays due to a storm, a dock strike, or a mechanical problem could play havoc with schedules intended to have a ship calling at each port on the same day each week. This was no minor problem: Israel’s Zim Line, which sailed from America’s Atlantic coast through the Suez Canal to America’s West Coast—the closest thing to a round-the-world container service in 1980—arrived within one day of schedule on only 64 percent of its trips and was more than a full week late one trip in seven. If shippers were to decide that a standard point-to-point service was more likely to run on schedule than one circumnavigating the globe, round-the-world ships might find themselves hard-pressed to attract freight. In the face of these risks, Sea-Land gave up its plans to sail around the world.
Two of its major competitors did not. One was Evergreen Marine. Evergreen, founded as a tramp company by the ambitious Taiwanese entrepreneur Chang Yung-fain 1968, had become a major operator across the Pacific and on the Far East-Europe route, undercutting conference freight rates to gain traffic. In May 1982, Evergreen ordered 16 containerships from yards in Japan and Taiwan at a cost of $1 billion and announced that it would run round-the-world services heading both east and west. The vessels, originally planned to carry 2,240 20-foot containers, were soon redesigned to hold 2,728. Chang called these vessels his “G-class” and named them accordingly: Ever Gifted, Ever Glory, Ever Gleamy. They would steam at 21 knots, fast enough that each of the 19 ports of call would see an Evergreen ship in each direction every 10 days. Evergreen’s ships would circumnavigate the world in 81 days east-bound, 82 days going west.17
The other competitor in the race around the world was an equally self-confident shipping magnate, Malcom McLean. In 1982, his U.S. Lines placed orders for 14
gigantic containerships. By building at Korea’s Daewoo shipyard, the company forfeited any rights to U.S. government construction subsidies but won the freedom to deploy the vessels where it chose, without government involvement. Each new ship could carry the equivalent of 4,482 20-foot boxes, half again as much as Evergreen’s G-class vessels. The ships were wide, flat, and utilitarian, designed—in the words of their architect, Charles Cushing—to look “much like a big shoe box above the water.” McLean’s strategy was different from Chang’s. His ships would circle the globe only in an eastbound direction, and they would do so slowly. McLean had learned from his mistakes with the speedy SL-7s, whose fuel bills ate up all their profits. The new ships were built for an era of expensive oil. They would conserve fuel by sailing at only 18 knots, taking longer than Evergreen’s vessels to sail around the world.18
McLean dubbed his new vessels Econships, because their fuel economy, along with the scale economies created by their enormous size, would produce the lowest cost per container of any ships anywhere. The ships alone cost $570 million. McLean’s new, publicly traded holding company, named, like its predecessor, McLean Industries, had no difficulty raising the money. The world was eager to invest with the founder of container shipping as he turned U.S. Lines into a “global bus service.”19
The profit potential of these new services was questionable from the start. Almost the entire container shipping industry was struggling: Seatrain Lines went bankrupt in 1981; Delta Steamship and Moore-McCormack Lines collapsed into the arms of U.S. Lines in 1982; Hapag-Lloyd sold its headquarters building for cash to repay creditors; Taiwan’s Orient Overseas holding was forced to restructure $2.7 billion of debt; and Sea Containers, which owned British ferry services as well as 15 containerships and a container leasing company, nearly went under. Matters grew even worse in 1984 and 1985, as Evergreen and U.S. Lines began their round-the-world services. New shipping capacity flooded onto the market. The available space for containers in the North Pacific rose 20 percent between May 1983 and May 1984, leaving ships from North America to Japan running almost half empty. Lloyd’s Shipping Economist reported “widespread rate cutting actions by carriers in search of market share.”20
The Box: How the Shipping Container Made the World Smaller and the World Economy Bigger Page 28
