Transportation Deployment Casebook/2018/Intermodal Shipping Container

The Intermodal Shipping Container
The humble shipping container has been a very influential transport technology greatly contributing to the growth of international trade and facilitating global trade, logistics and supply chain development.

Standardised containers come in several varieties but they are typically an enclosed rectangular box with steel sides and roof, timber floor, and doors at one end. The most common shipping container now is a 40’HC, (forty-foot High Cube, which is not a cube) which is 12.2 m long, 2.44 m wide, and 2.90 m high, though the next most common are the 40’DV and 20’ DV which are both 2.59m high, with the 20’ only 6.06m long. The 20’ is the basis for a common standard of measurement, the TEU (Trailer or “Twenty Foot” Equivalent Unit), used for ship and port capacity and container fleet capacity. For example, each of the 40’ versions mentioned here are considered as 2 TEUs, and the Port of Shanghai handled 34,516,000 TEUs in 2015[1], and OOCL Hong Kong, rated the highest capacity container ship in 2017, has a carrying capacity of 21,413 TEU [2].

Advantages
The container’s key features are its “corner castings” on the 8 corners that are part of a frame rigid enough to allow it to be securely stacked on other containers. The container can be easily mounted on standardised trailers for road transport, a railway car, or most importantly – stacked within or on a container ship ten high. Further, a simple twist lock connector can easily create secure links between stacked containers, top corner beneath to bottom corner above, allowing a stack to be a rigidly connected structure. Though longer length containers exist, their connectors are located such that they stack on a 40’ with equally overhanging ends. The corners similarly also allow easy connection to cranes or booms which greatly facilitates loading, unloading and transferring between modes – hence “intermodal” containers. [3]

Market
These advantages and the spread of standardisation between modes, the ease of creating special containers within the rigid connecting frame and corners, and continuing innovations in ancillary container handling technologies, have seen the container capture most of the shipping market for non-bulk goods over the last 60 years, and overwhelmingly most of the value of seaborne trade. Containers are ubiquitous in trucking, and rail, and non-bulk freight that is not on the last-mile is probably in an intermodal container. By delivering cheaper, faster and more secure shipping, trade growth and globalisation were assisted which resulted in a virtuous circle positive feed-back loop between containerisation, and international trade. The scale of operations in seaborne shipping, furthered developments in ground transport, and allowed logistic concepts such as “just in time” delivery, distribution centres, and “global supply chains” to develop furthering globalisation.

Before the Container
Prior to containerisation, cargo ships had been developed for bulk liquids -tankers – from 1863, and dry bulk commodity cargos – bulkers - such as iron ore and coal, or for grains. The first recognised bulk carrier dates from 1852. By contrast, other cargo, known as “break-bulk”, was piece loaded into cargo bays onboard by dock workers and longshoremen well into the twentieth century. Block and tackle, cranes and nets and other machinery would assist the process but at each stage of the transfer process extensive human labour was involved. Whether by truck, train or carriage a load might arrive at a wharf, and men would remove it and store it about the port somewhere – transit shed, warehouse, or even dockside. Random boxes, cartons, crates, sacks, baskets, casks, barrels and all manner of non-packaged goods would be stacked together. Loads might be transferred through the process of stacking up a pallet then hoisting that into a cargo hold by crane where another team of men would remove the goods and arrange them by hand around the cargo hold, or men might simply carry goods up and down gangplanks [4]. Moreover it was generally more common for the ship’s own lifting gear, rather than the dock’s, to load the ship [3]. Whatever the nature of the goods, whatever mode brought it and whatever the size of the ship an incredible number of men and man-hours were involved and a ship might spend days, or even weeks, in port being unloaded and loaded.

Despite the emergence of iron ships powered by steam and the slashing of journey times, and increases in reliability of schedule over the 19th and early 20th century, the handling of break-bulk cargo in port remained much the same as it had been for centuries, even with the development of fork-lifts and other mechanical accessories. The amount of labour involved significant time and money costs, and losses to theft were common. Attempts had been made to overcome these issues with various schemes such as pallets and smaller containers but organised labour or the stevedoring companies themselves would ensure delays through obstructionist work practices that produced even worse efficiency and the shippers tended to give up.

On the other side of the dock there had also been a great deal of innovation – railways had developed into a mature mode servicing hundreds of lines emanating from the world’s ports and efficiently linking both inland cities and resources to the coasts of lakes, seas and oceans. Various containerisation projects were used with varying results particularly in railway goods handling, even from the first railways. Also in the early 20th century use of the automobile started to grow, assisting road development and allowing trucks to become a feature of goods transportation. American railways had responded to this with the so-called “piggy-back” system that by various means put a truck trailer onto rail cars.

These innovations all made transportation of goods more and more viable, and helped drive the growth of ports and shipping before World War I, and to an extent hid the inefficiencies of cargo handling. WW1, the Great Depression and WWII interrupted international trade significantly, though significant technological developments had occurred by the time peace broke out.

The Innovator
World War 2 had been a war of logistics. The vast scale of the conflict, and the vast distances involved, particularly in the Pacific, had seen technologies rapidly developed particularly automobiles and aircraft, but in rail handling as well. While the shipyards of the US had produced an extraordinary number of ships, including over 2,700 of the iconic Liberty Ships, cargo ships capable of carrying 10,000 ton loads [3], the post WW2 maritime trade in the US was stagnant [4]. Eventually large numbers of standard tanker and cargo vessels from the WW2 logistics fleets made their way into the fleets of shipping lines. Trucks were grabbing market share in the US, and their technology was much improved due to war experience. The coastal shipping of the US was much reduced from its pre-war levels, even into the 1950s [3]. While not particularly profitable maritime trade was highly protected, subsidised and not particularly in any rush to innovate [4].

It took an outsider to bring disruption to the docks, and with it the entire freight industry. Malcolm McLean was a trucking magnate who had grown his business in the late 1940s as truck transport thrived, and the railways and shipping stagnated. All modes however were tightly controlled by the Interstate Commerce Commission (ICC) which had stability as its main goal [4]. McLean had long thought the docks were a major source of time being wasted for trucks in unloading, and was very focussed on cost cutting due to the “set rate” freight regime set by the ICC. However, it was not until 1955 when his goal had been to beat trucking competitors through price competition through use of the lower allowed rates available to east coast shipping, that his attempts at disrupting trucking competition eventually led to him forcing innovation on the entire freight transport industry [4]. While originally pursuing a truck rolling on, rolling off style of transport [5], McLean decided the undercarriage and trailer was a waste of space and would prevent stacking and wound up wanting stand-alone containers that could just be lifted off trailers at one end and put on trailers at the other end.

While steel boxes were already available they were not at a size that would create genuine cost-savings [4]. Pushing his own staff at his recently acquired Pan-Atlantic Steamship Company to come up with something, he was introduced to an engineer, Keith Tantlinger, who had already designed and sold, on behalf of his employer, 200 of his 9 m long Aluminium containers for use on barges but that were able to fit on a chassis to be pulled by a truck. Tantlinger’s containers were accepted by McLean. Tantlinger, enticed to work for McLean, eventually also had the job of convincing the Coast Guard that the containers were safe on-board ship, and he developed the spreader bar that enabled the dockside crane operator to “grab” the corners of a container, pick it up and move it and release it without another worker’s involvement [4]. After delays due to competitor appeals to the ICC and Coast Guard approval, McLean finally got away his first converted tanker loaded with containers on April 26, 1956 – at the rate of a container every seven minutes – and underway from Port Newark, leased and converted by McLean, bound for Houston [4].

The innovation here was not the container itself, as others had thought of them, built and sold them, used them on routes of both rail and sea, and attempted to promote them. The current International Container Bureau (BIC), which assigns ID numbers to the now ISO standardised containers, had been established in 1933, [6] over 20 years earlier, and was tasked with promoting the use of standard sizes though these were not as large as the eventual shipping containers would be. Nor was moving a container from a ship to another mode, nor the use of a special crane, using a privately leased port, nor outwitting both regulators and organised labour, unheard of before. Rather, McLean saw the whole system as a mechanism to move cargo, and that everything in the chain had to be improved towards that end, and in putting together the company structures, a specialised port, getting terms from labour, modified ships, the containers, and identifying suitable customers who would ship enough volume with him for long enough to be considered commercially successful, he changed shipping economics – despite having had no previous experience in ocean shipping [4, 5]. He was rewarded with substantial cost savings, that led to both competitors’ appeals to regulators, emulation of his techniques, growth in competition and substantial growth in available markets. The world economy was rewarded with a fundamentally cheaper and more efficient freight system that would become much more of both in the coming decades as it scaled and standardised, and would profoundly impact global trade patterns and with them, national economies becoming more trade focussed and even trade based which further promoted the scale of containerisation [3].

Early Markets and Enhancements
While McLean had entered the world of seaborne freight in a rush, with quick generation and abandonment of ideas, and doing the best that could be done with what was to hand, there were several innovations that Tantlinger quickly moved on.

McLean counted on pent-up demand and pushed hard to increase operational capacity. A longer container, more containers to a stack, converting break-bulk ships rather than tankers (which were still useful as tankers), converting holds to optimal dimensions for containers, and widening hatches were easily identified improvements. Creating cells for the containers to sit within below decks that would be would be wide enough to manoeuvre the containers in with the crane but not so wide that the container could shift was more difficult [4]. McLean’s Pan-Atlantic had converted C-2 ships (a standard US cargo ship of WW2) with quadrupled capacity in 1957, just a year after their first effort with converted T-2 tankers. This meant crane innovation was needed to keep the turnaround time for the bigger capacity ships as low as for the initial ones, and since ports would not necessarily have the right cranes, ships needed their own cranes too. Stronger corner posts to allow higher stacking, producing refrigerated units, door improvements and the twist lock that allowed for easy locking and unlocking at the corners were all quickly adopted.

The market power was immediately apparent. A typical C2 break-bulk cargo ship, would take 150 longshoremen working four days to load and unload, perhaps incurring $15k in labour costs alone. McLean’s first C2 conversion – the Gateway City – which commenced operating in October 1957, could be turned around by a 14 man team in just over eight hours, for a labour cost of $1,600 [3].

Other more established shipping lines were impressed but were reluctant to invest until some of the concepts shook-out and standards emerged. One operator Matson, took on the Pacific market in a very deliberate and scientific process that particularly developed the shore based crane [4].

One of the key early markets for McLean’s operation – now Sea-Land – was Puerto Rico. Initially out-manoeuvred he was eventually able to virtually build a monopoly and a virtuous circle was created that greatly assisted in the economic development of the island that in turn helped improve Sea-Land’s profitability. This was achieved by slashing cargo rates and stabilizing delivery schedules due to the productivity gains achieved by containerising. [4] This was a prelude to what would happen in Asia over the coming decades as developing economies shipped their manufactures to international export markets.

Overall, the first 10 years after 1956 were characterised by solid but relatively linear growth, as Pan Atlantic/Sea-Land and Matson added higher capacity ships, converted ports as they added new domestic service routes, encouraged organised labour compromise, and optimised their systems. Potential competitors investigated, but tended to adopt a wait-and-see approach during this first 10 years, but participated in the standards development process, evaluated their own opportunities and made plans – chastened by the failure of Grace Lines with Venezuela (below in Policy). Eventually in 1966 two other lines and McLean initiated North Atlantic services between the US, Europe and the UK. While initially McLean had started point-to-point on dedicated lines, with the move to international lines there would develop more of a hub to hub system, with smaller feeder lines utilising smaller ships to service other ports – much like the trucking system. Containerisation exploded quickly after that, and by 1969 there were 199 purpose-built container vessels under construction world-wide [3], and by 1971, just fifteen years after Pan-Atlantics first US domestic container service, and five years after the first dedicated transatlantic container service “all cargo on the transatlantic routes out of the US that could be containerised, had been” [3].

Policy in the Birthing Phase
Initially, policy was largely directed against containerisation, at least in the US. Containerisation, and all forms of freight consolidation such as palletisation were controversial to organised labour and thus inherently political. Physically each mode, and within each mode each operator, faced unique conditions or had significant other investments that it wanted to work with or avoid threatening and so there was contention to work through towards standard dimensions. Finally, the infrastructure of each mode involved a wide range of regulation, law and governing jurisdictions.

The ICC kept strict lines of separation between modes, and there were strict rate and competition controls. Ironically it was this rate-control environment that pushed McLean into containerisation in the first place. A group of railway companies complained to the ICC about the competitive effects [3], and the Coast Guard were involved over concerns about the safety for the crew of carrying containers on deck [4]. In the US case the railways were keen to defend the use of their own rail cars and their own “piggyback” system for hauling truck trailers.

Ironically the ICC whose initial role had been to prevent the market being exploited and to ensure common carrier standards was now being used to keep competitive forces out of the market and was thus protecting inefficiencies and limiting innovation. ICC rules also forced McLean, eventually to remove his family from running their own trucking firm, and he had to use other trucking firms for removing containers from ports.

The Jones Act, still relevant today, enacted in the 1920s also kept coastal shipping as an all American affair within the US [3], which simultaneously provided relief from foreign competition but also limited the ships that could be used and imposed higher crew costs.

However, McLean also – at least on the financial level, and solvency is an important characteristic for an innovator – had an ability to find useful policy assistance, when it came to plans to convert and/or buy ships, without using the subsidy plans that would have controlled routes. Whatever policy difficulties his succession of corporate entities faced, they applied to his competitors as well. Additionally, he benefitted from the tension between the New York Port Authority and the City of New York, in being able to establish himself at their new development at Newark, with much better road and rail access than the New York city docks, and the Authority benefitted from his success. In terms of engineering innovation there was little initial intervention though there was the question of sea-worthiness and crew safety, with the American Bureau of Shipping (ABS) and US Coast Guard being involved in sea trials of McLean’s converted T-2 tanker, the Ideal X [3]. However, in switching to longer containers than the original 9 m containers when Pan-Atlantic switched to its specially converted C-2 ships, which had container cells below deck, the length chosen was 10.67 m (35 feet) because that was the largest length that would fit on the maximum length trailer allowed on Pennsylvania roads, not far from New Jersey where his dock was, and McLean would stick with this length into the 1970s [3].

By 1958, though, the American Standards Association already had a panel to investigate a container length standard and worked with the ISO over the coming years. Bizarrely, the two lines actively utilising and growing the container business were largely non-participants, and their used lengths would not win out. Eventually it was determined to work with lengths of multiples of ten feet (3.05 m), even though no one was using them at the time. Only twenty and forty feet emerged as seriously used lengths, until more recently, though many different lengths would come and go in the interim before the ISO managed to publish its first full draft in 1970 [4]. It was recognised that the corner castings were crucial, and much debate ensued about which system to adopt. McLean, in 1963, was persuaded by his old employee Taitlinger, then heavily involved in the standards process, to freely allow use of his organisations patents to the “corner casting” which helped set containerisation on its path of inter-changeability [4], and gradually the container fleets moved towards the standards. Although large established fleets of non-standard containers had enough value to only gradually be changed out, new entrants accepted the standards as locked in for a few decades, though extended road trailer lengths routinely challenge the lock-in.

The main policy threats were tightly controlled competition and anti-collusion rules, along with labour concerns about the threat to employment, and where organised labour goes so must government policy – either for or against. Although McLean was able to strike a deal with both the Teamsters (trucking) and Longshoremen (dock workers) organisations [3], because he was apparently creating jobs at a new site (Newark), labour concerns would prove to be a consistent bugbear internationally through the coming decades as ports confronted the decision to containerise. In 1960 Grace Lines tried to set up a service to Venezuela only to be completely cruelled by local labour refusing to unload, and no solution was successfully negotiated, and their plans abandoned [3]. Organised labour in various ports and countries would respond negatively to containerisation, including Australia, but would often find themselves swept aside as a competitor port in another jurisdiction seized the opportunity, such as Oakland, or just through happenstance and lack of relevance, such as Felixstowe in the UK, and by its nature container shipping would go on to support fewer, larger ports. Early containerisation however, proved not to be a guarantee of long term success for all ports [7]. Ports and the jurisdictions with interests in them whether city, provincial or national would develop containerisation policies but success could be elusive.

Growth of Containerisation and Policy Development
As containerisation captured the break-bulk market it also had profound impacts on the economies able to utilise the cost and time savings to freight – as seen in Peurto Rico in the late 50s and through the 1960s. The initial impact, relatively – it takes a few years/decades - is that existing production can be more quickly and securely transported for less and to a more reliable schedule. This however changes markets, and the less friction in the transport system the more local comparative advantages come into play in international trade, the less need there is for industries to be physically close to their markets and the more competition is available for sourcing components or finished goods. Initially, outside the ICC, containerisation had support from politicians who saw the potential for American goods to be more easily exported, while the ICC saw it as disruptive and labour saw it as the major threat to jobs. Importing, however, also became much simpler.

Standards shook out, not just in dimensions, but also in registration, identification and marking – all covered by ISO standards. As shippers moved to ISO standards, it become easier for manufacturers of ancillary products and technologies to scale, and as internationally trucking and rail followed shipping there were ever bigger markets to reach. While the move into trans-Atlantic operation which initiated explosive growth was largely private lines as the driving force there were always government actors – some state-owned shipping lines for example, ports, and railways to be engaged. Also in 1966 one of the important Pacific developments was McLean’s Sea-Land securing important military contracts in providing container logistics for the Vietnam war effort. This gave him free capacity to develop on the return trip which provided greater profitability when he secured some of the nascent Japanese export market.

Merchant fleets have moved away from government ownership, and even being flagged to their country of ownership, and so it has been mostly the private sector that has led the way in expansion of fleets, building ships with more TEU capacity, and many container ports are privately held. For example, although it was US operators, and effectively the US merchant marine that lead the way a long-term result was the US merchant marine fleet losing international relevance. Simultaneously though many ports remain nationalised, and even some ship lines, customs are still publicly controlled, many railways still government owned and common carrier rules apply. Many policy trends however have enabled the growth in the containerised freight market.

The Asian experience – Japan, Korea, Singapore, Hong Kong, and China containerisation development was either government controlled altogether or government led. Larger trends of de-regulation, certainly in OECD nations, have enabled capital to move freely and more responsively to market conditions. Free trade agreements, the expansion on the WTO, and the development of trading blocs has helped make the reduced transport friction that containerisation has brought far more relevant to international freight than it otherwise would be.

Containerisation was a boon to Asian economies, both to Japan and the tiger economies, but also eventually to the huge cheap labour pool that was mainland China. Initially China was out of the picture, but various policy changes from Nixon in 1971 through to its own market reforms of the 1980s and after with eventual WTO access led to it being able to export its cheaply produced manufactures to the world, while the “bulk” market benefit from its imports, and that continued and expanded the threat to American manufacturing that Japan represented in the 1980s. Similarly, the continuing refinement of the shipping container and its handling, was both encouraged and demanded through increasing international co-operation on trade, free trade agreements and ISO adoption in new markets. There has been over the last 20 years a massive shift to an Asian and especially Chinese presence in the busiest and biggest container ports, usually measured by TEUs processed. Including Hong Kong Chinese ports occupied seven of the top ten port container volumes in 2017 [8].

Maturity; development, market change, competition and policy
In 2016 the container revolution started by McLean turned 60 years old. While many things continue to grow, such as port and ship size, there are some physical constraints and other modal developments that might suggest peak containerisation is near – though physical constraints have been pushed through before. In 2008 Notteboom and Rodrigue suggested that maturity might be near despite continuing volume growth, but they noted that many signs of maturity were already present, especially the increasingly marginal technical improvements[7], a trend which has only continued since, though higher TEU capacity ships continue to be developed and built. For example various forms of route organisation abound – EG triangle, pendulum, butterfly, conveyor belt, line-bundling, end-to-end – that allow lines to fill various shipping niches at optimal efficiency [7]. The processing chain for containers is constantly improving incrementally. While crane sizes might be topping out, there is the typical mature squeezing for efficiencies in their operations including multi-box operations, automation, and ever more powerful computerisation. Port design, stacking and storing optimisation software, container tracking and scanning are other means of getting the most out of containerisation and given the scale of containerised freight the slightest improvements of even fractions of a second or cent can have high value [4].

While many ISO standards apply and about 90% of all units are the three main types of dry van, the fact that the basic principle is the twist lock corners on a robust frame mean there are a bewildering variety of “containers” compliant with the broader system but which are not “boxes”. Aside from the early development of the refrigerated unit there are specialised containers that put a variety of tanks inside the core frame, side open, and top open containers, longer (and wider) containers, and containers that are suitable for carrying smaller quantities of otherwise bulk materials. Again, this is typical of maturity as specialisation to capture niche markets occurs.

A threat has been the opacity of containers – while uniquely identified and potentially with a manifest the only way to know what is inside is to inspect them. This provides smuggling and terrorism opportunities which policy makers want to address. Physical inspection by humans is time consuming and delaying and scanning technologies are being utilised and developed.

Quantitative Analysis
Unfortunately consistent data across the history of the container is difficult to obtain, in part due to the time it took for it to standardize. The standard measure for handling containers is by TEU but there are several problems, particularly in reference to moving and handling empty containers. Recent port handling statistics are well developed, but are not necessarily useful for value of contents or distances travelled. Weight is not a particular valuable measure as it does not measure container numbers nor value, but it does allow comparison with other shipborne trade and potentially bulk rail traffic. The 2017 Report on Maritime Trade by the UN Conference on Trade and Development published a figure showing a comparison of maritime modes tonnage shipped dating back to 1980 and up to 2016. Unfortunately this misses the earliest 24 years of the mode, and limits the ability to determine the top of the probable S-Curve since it is does not reveal the early part of it.

Analysis of the data below suggested an upper tonnage of 3125 Million Tons, and that about half way through 2013 was the inflexion point for the potential S-Curve. The graph does suggest that the rate of growth was increasing from 1985 through to five years ago. If 2013 or similar was indeed the inflexion year, then the growth ahead for intermodal containers will remain strong for another decade or more but the rate of growth is already starting to decline. Also while we may not have earlier data we do know the starting year quite well for dedicated container shipping - 1956, so it took 24 years to reach 100 M tons, and only another 8 years or so to double that (1988 for 200M), but then doubled again in the next 8 years (400M in 1995) and then again in the next 8 years (800 M tons in 2003) while taking 10 years to double again suggesting the rate of growth was indeed slowing. However, the GFC of 2008 did slow the global economy for several years.