After several years of significant growth, the North American intermodal network once again finds itself approaching capacity. As the industry contemplates how it will handle current volume levels, users of intermodal services are on edge because there is a widespread fear that the railroads will not be able to handle any significant intermodal growth next year, and that they will attempt to manage supply and demand by raising rates.
The slightest hiccup in this network reverberates around the world and disrupts supply chains near and far—witness the impact of port congestion on the U.S. West Coast, and of two consecutive winters of abnormal cold and snowfall that literally stopped trains in their tracks. This situation makes it imperative that the railroads find a solution that could accommodate both current and future intermodal volumes.
The traditional response has been to build more infrastructure—more trackage, more intermodal facilities, more container terminals with on-dock rail. But building more of what worked in the past will not solve problems that are rooted in quickly changing economic and trade patterns. Moreover, building additional infrastructure is incredibly expensive, and land for expansion is simply unavailable in the congested urban areas where most container terminals and rail yards are located. Furthermore, the completion date of any new infrastructure would likely come well past the point of gridlock.
Infrastructure isn't the only thing holding back capacity expansion. As this article will explain, the main players in the intermodal system—the railroads, the ocean carriers, domestic shippers, the container terminals, and the drayage carriers—have conflicting economic and operational mandates. What benefits one works to the detriment of another, creating a disincentive for profit-making companies to implement significant change.
I believe that it is time to rethink North America's international intermodal network from the water's edge inland. If the North American intermodal system cannot accommodate more volume, it will have a negative impact not only on importers and exporters but also on economies throughout the Pacific Rim and beyond. But this is not solely an international issue. The container-based domestic intermodal system is subject to the same capacity concerns, and therefore the shared domestic and international network must also be considered.1
This article will look at why the intermodal system may be incapable of handling current and future demand. It will also examine two major nodes, Southern California and Chicago, as well as two intermodal terminal alternatives that could help to provide the increased capacity that the industry desperately needs for both domestic and international container movements.
The impact of rail economics
Intermodal transportation is an asset-based, network-operating business. In theory, it is an integrated system in which railroads, ocean carriers, container terminals, and drayage carriers cooperate to create a smoothly operating service with efficient handoffs, sufficient capacity and equipment available when and where needed, and clear communication. That is an ideal that is rarely achieved, and this shortcoming—much of it caused by the conflicting interests of the four main players—is a major reason for the challenge facing the intermodal system when it comes to handling future growth. To get a clearer picture of the contributing factors to the intermodal growth problem, consider the conditions and challenges confronting three of these entities: the railroads, the intermodal operators, and the drayage carriers.
Among the myriad problems facing intermodal operators, the ability to provide reliable, truck-competitive transportation may be the most daunting. Truly competitive intermodal transportation must provide an all-in, door-to-door transportation service that is less expensive than and just as reliable as the comparable truckload product. However, many factors are conspiring to make that difficult for intermodal operators to achieve. Among them:
Over the past 15 years the rail intermodal network has grown by concentrating train service into end-to-end unit trains that operate between large intermodal terminals. Historically, the preference had been to run a train with multiple origins and/or destinations that involved "block swapping." Trains would setoff traffic while en route, with a different train picking up and taking the block of containers or trailers on flatcars to the destination. This was tolerable when intermodal volumes were smaller. However, a single, late inbound connection could delay numerous outbound departures.
Most intermodal networks today try to avoid intricate or complex loads, which threaten to adversely affect existing volumes and destroy a train network's density and economies of scale. Railroads prefer to run solid "unit" trains between two points, ideally without intermediate stops, which consume network capacity. This presents the lowest network operating cost and the simplest and most reliable service, but it requires significant and steady, balanced volume. In an attempt to achieve that, railroads must often hold outbound on-dock train segments in space-constrained rail yards adjacent to marine terminals until they have sufficient volume to create a sizable block for a single destination, a practice that exacerbates congestion and delays.
At the same time, inter-railroad connectivity has become an especially acute problem. Rail-to-rail interchange (usually between eastern and western railroads) traditionally has involved highly desirable long hauls and is accomplished through major, established gateways such as Chicago. While "steel-wheel" (rail-to-rail) connectivity is still possible, it has become the exception. Most inter-network connectivity in recent years has been accomplished by "rubber-tire" (truck to rail, or rail to truck) interchange.2
Ocean carriers' intermodal movements usually involve at least one major port where railroads operate daily service. For domestic intermodal operators, the existence of rail intermodal service cannot be taken for granted, even when the distances involved seem amenable to rail transportation. Although intermodal has been successful with lengths of haul greater than 700 miles, in most cases, shorter lengths of haul have been confined to a single railroad. For example, the distance from Minneapolis, Minnesota, to Pittsburgh, Pennsylvania, is 870 miles and should be a sufficient length of haul for intermodal, but because it requires a combination of two 400-plus-mile moves, on two railroads connecting through Chicago, there is no integrated rail service offering. This lack of short-haul intermodal service makes it complicated and expensive for international operators to move between many inland origins and destinations.
The inter-railroad connectivity problem impacts more than current traffic; it is a key determinant of future intermodal growth. If railroads could make shorter connecting hauls cost-efficient and profitable, it could open up a whole new intermodal market that does not exist today.
The impact of carrier alliances
Over the last 15 years, ocean carriers have concentrated their focus on conveyance economies of scale. Vessel sizes continually surpassed previous estimates of maximum size. The 4,400-TEU (20-foot equivalent unit) C-10s operated by APL, the world's largest container ships when they were introduced in 1988, are now dwarfed by the 18,000-TEU MÃ¦rsk McKinney MÃ¸ller, which was deployed in 2013. Yet these linehaul efficiencies are commonly offset by problems elsewhere:
Ocean carriers want bigger ships so they can reduce their linehaul cost, but today's giant vessels are actually raising costs and creating bottlenecks on the landside portion of a container's journey. Back in the earlier days of on-dock rail, intermodal trains would essentially function as an extension of the ship; the priority was running trains with blocks of containers for specific customers to a particular destination. For example, Toyota and Honda would have enough inbound volume on a single ship to justify a container train to the Ohio Valley.
That has all changed. In the past few years, trans-Pacific shipping has been transformed by monolithic vessel alliances and the deployment of mega-ships. The scope, scale, and complexity of these arrangements threatens to overwhelm the on-dock infrastructure that has been developed over the past 25 years and increasingly looks like the Maginot Line in 1939: impressive infrastructure that is unable to fight a modern war.
Nowhere are these problems more apparent than in Southern California. Transloaded imports are estimated to represent 70-90 percent of domestic rail movements from California—the "mother river" of North American intermodal traffic. Southern California, then, is unique in its influence on both the international maritime and domestic intermodal systems.
With the advent of increasingly larger container ships and the rise of vessel alliances (with as many as six distinct carriers sharing a single vessel), railroads serving the U.S. West Coast must contend with not 3,500 but 12,000 or more TEUs, and not one shipping line but as many as six lines. That means railroads and container terminals are dealing with the equivalent of six 2,000-TEU ships. This, in essence, creates a sort of transportation "Tower of Babel"; the railroads cannot run a train with small numbers of containers for the many destinations required by the multiple carriers sharing a single ship. (See Figure 1 for a comparison of intermodal train plans before and after the advent of carrier alliances.)
The carrier alliances further complicate train loading by exponentially increasing the number of contracts and commercial priorities the railroads must deal with. Making matters worse, alliance members may operate in multiple terminals within a port, so containers and chassis coming off a single ship may have to be returned to different terminals, adding to congestion, delays, and dissatisfaction on the part of the local draymen. It's as if every airline landing in Los Angeles had its own airport.
The impact of drayage issues
Intermodal drayage—the trucking services that move containers between the rail yards or container terminals and the shipper's or consignee's door—plays a crucial role in intermodal service. To compete with door-to-door truckload service, intermodal must replicate that service. To do that, intermodal providers market door-to-door service built around rail intermodal movement. By bundling pickup and delivery, equipment, and rail movement, these providers offer a one-invoice service similar to that of a truckload carrier. The drayman provides pickup and delivery services between the customer locations and the linehaul terminal. (Drayage, by the way, is a tractor-only service. An equipment provider or the customer itself supplies the trailer or container.)
Intermodal success also requires seamless transitions between service providers. While all members of a relay team are important, the first and last runners are critical. The first runner establishes position. The last runner, or anchor, is tasked with securing the win. Similarly, the drayman, who fills the first and last positions—pickup and delivery, the only two stages visible to the customer—is essential to the intermodal customer's satisfaction.
But intermodal drayage is in crisis and has imposed a ceiling on intermodal growth. The intermodal network of the future needs to reflect these new realities.
Everyone assumes that intermodal service is cheaper than truck, but that's not necessarily true. On a cost-per-mile basis, rail linehaul is a fraction of the cost of truck. But when drayage is included, the inland expense for international container shipments can be almost as great as the total ocean linehaul costs. In some cases, inland costs may even exceed the ocean transport revenue. For domestic shipments, combined rail and drayage expense can be higher than existing door-to-door truck rates.
One reason inland costs are so expensive is that railroad intermodal, which travels over a fixed network, faces a circuitry () problem.
As shown in Figure 2, in an ideal world:
As shown in Figure 3, in a non-ideal world:
Even in an ideal world, there is a challenge. As seen in Figure 4:
The obvious solution to this challenge is to reduce the drayage distance and likelihood of unfavorable circuitry by serving more intermodal terminals. Put another way, in order to be cost-competitive with truck, drayage has to minimize its length of haul. That requires greater scope—getting as close as possible to the door-to-door origin and destination. Thus, although the trend has been for fewer, larger intermodal terminals, there is a need for more, smaller terminals that are closer to the customer.
Drayage providers face another serious problem: a labor shortage. In the United States, over the past 10 years enhanced regulatory requirements, along with demographic changes, have reduced the pool of available truck drivers. For example, the development of the national commercial drivers license (CDL) eliminated the ability to maintain multiple state licenses. In the past, a driver who lost a license issued by one state could still drive using a license issued by another. Today, if a driver loses a license, he or she is out of work. Mandatory drug testing and a minimum age of 21 also eliminated many prospective drivers. Changes to hours-of-service (HOS) regulations reduced the amount of allowed driving hours, and enhanced medical oversight diminished the pool of acceptable drivers. The imposition of electronic driving logs and truck-speed governors will likely further shrink the pool of drivers because drivers will be restricted in the number of hours or the number of miles they drive.
These regulatory changes were imposed with the aim of improving safety, and the industry continues to debate their effectiveness and necessity. Regardless of one's opinion on the matter, the fact is that since drivers are paid based on the amount of work they perform, any regulatory change that reduces the number of overall drivers, or the number of miles that the remaining drivers can cover, will increase scarcity. Scarcity will become crisis if demand continues to grow.
At one time, experts believed that such requirements would benefit intermodal because they would disproportionately affect truckload carriers. However, the impact has been more widespread; drayage companies now find themselves competing with more traditional motor carriers for drivers and often coming up short.
Drayage offers drivers the possibility of regular hours, but it does come with other challenges. Intermodal, with its potential terminal congestion and train delays, can unpredictably impact task times, with the result that drivers often are forced to squander some of their on-duty time sitting still. Highway and contract logistics, by contrast, are door-to-door moves, with no intervening terminals.
Some drayage carriers are reporting that owner-operators, looking to be classified as employees, are complicating the business. In September 2014, the Hub Group was forced to convert 350 owner-operators to company employees in California. A few years ago, Hub might have shut down its California operation and purchased drayage from third parties. Hub's 2014 action may signal an acknowledgment that the company could not presume sufficient purchased capacity under any circumstance and should deal with company employees rather than take chances with outside providers.
Two possible solutions
Network design (and related infrastructure) historically has been more about engineering than economics. As a result, robust design frequently becomes economically obsolete before the infrastructure itself physically deteriorates. This phenomenon is visible today in marine terminals designed for single lines and smaller vessels, warehouse clusters mooted by the relocation of manufacturing and consolidation, carload docks designed for 64-foot (not 72-foot) boxcars, and rail intermodal facilities that are no longer able to achieve critical mass.
There is no obvious systematic or uniform response to these networkwide challenges. Moreover, Southern California's unique set of capabilities—terminal scale, rail scale and scope, warehouse and DC scale, and a large local population base—exist nowhere else in North America. A change in both the on-dock model as well as the supporting intermodal network model, therefore, is likely the best solution.
Two intermodal initiatives, representing two distinct network concepts, provide models for accommodating intermodal growth: BNSF Railway's proposed Southern California International Gateway (SCIG) and CSX Transportation's (CSXT) Northwest Ohio (NWO) intermodal terminal in North Baltimore, Ohio.
Since 1980, when BNSF predecessor ATSF declined to participate in Southern California's Intermodal Container Transfer Facility (ICTF), that railroad has sought to expand its capabilities in the San Pedro Basin, which includes the ports of Los Angeles and Long Beach. BNSF has tried for years to build SCIG on Port of Los Angeles land, but the railroad is engaged in a long-running battle with local activists who oppose the facility, and it is uncertain when SCIG will be built.
The proposed SCIG is a near-dock facility that would complement on-dock facilities developed over the past 30 years. While both on-dock (self-contained within an existing marine terminal) and near-dock (a stand-alone facility adjacent to marine terminals) facilities perform intermodal rail transfers, these terminal types operate under different network constraints.
The scheduled on-dock network scope is currently restricted. At present, BNSF offers on-dock rail service to six cities, whereas its off-dock Los Angeles Intermodal Facility (also known as "Hobart," located a few miles east of downtown Los Angeles) offers service to over 20, according to BNSF's 2015 container service guide (see Figure 5).
Although SCIG would offer the advantages of a near-dock facility, it is a product of a status quo mentality. It is a traditional engineering solution focused on the infrastructure's expected physical life, rather than a solution that is based on economic analysis and is focused on expected useful life.
Over 2,000 miles east of Los Angeles lies CSXT's Northwest Ohio terminal. CSXT officials describe NWO as the railroad's third Chicago intermodal terminal, even though it is located 265 miles east of that city. NWO was designed as a network hub, rather than as a traditional origin/destination terminal; it feeds numerous terminals on (and off) its network. As shown in Figure 6, it currently offers direct connectivity to almost 30 intermodal terminals.
NWO offers a new way to accommodate scope and scale, all while providing additional Chicago capacity. The facility solves the scheduling critical mass problem by utilizing a classic "hub and spoke" system, which concentrates volume through NWO and then runs trains between it and other network points.
NWO accommodates business between western railroad points and almost 20 eastern locations by allowing for a fourth option that frequently is better than the first three, traditional options listed below:
As an alternative to SCIG, if the western cargo origin is a San Pedro marine terminal, then the terminal could easily load a single train destined to NWO.
Option #4 could have a significant operational impact on the national intermodal system. Moreover, CSXT believes that NWO makes immediate financial sense for two reasons. First, rather than increasing complexity—always anathema to customers—NWO simplifies intermodal container shipping by eliminating the need to use Chicago as a transfer point.4 And second, smaller-volume terminals can be supported by a single, daily train that concentrates all volume through a single, intermediate focal point.5 This, in turn, supplies service to more terminals, thus allowing more proximate drayage service and making intermodal more competitive, without destroying any linehaul economies. The hub-and-spoke network generates economies of both scope and scale.
Just as NWO provided Chicago expansion outside Chicago, could an NWO-type facility outside Southern California offer an alternative to the proposed Southern California International Gateway? The current generation of marine terminals utilizes on-dock facilities that were designed as if they were near-dock operations. An import container is discharged from the vessel and comes to a "point of rest" (either on a chassis or in a stack) within the terminal. The intermodal operation is completely distinct from the vessel operation. The imported container is subsequently rehandled from its point of rest in the terminal and loaded to the rail car. Export containers are handled in the reverse manner.
Ideally, the future on-dock rail operation would be a shuttle-train operation adjacent to the quay, and the vessel crane would move a container directly to or from the rail car. This could eliminate the intermediate point of rest, thereby greatly accelerating terminal throughput and increasing terminal capacity without requiring additional infrastructure. The amount of space required for on-dock intermodal could be reduced by 60-90 percent, and space available in some terminals could increase by 140 percent. Not only would the resulting operation be vastly more flexible and fluid, but it also would generate some of the additional terminal capacity that is necessary—but not readily available—to handle larger ships.
This envisioned shuttle train would depart as a unit train in a faster, more efficient manner. Once inland, shuttle trains from the 13 San Pedro marine terminals could be reworked into unit trains destined throughout North America. Network hub locations as far as 1,000 miles away (such as BNSF at Clovis, New Mexico, or Union Pacific at Santa Teresa, New Mexico) could still generate the intended scope and scale economies.
The network hub could also be located in California's Inland Empire region, which is home to numerous warehouse and distribution centers. The attraction here is one of opening the ultimate shorthaul intermodal market. From San Pedro (using the railhead in Wilmington, California) to Colton, California, an Inland Empire city served by both BNSF and Union Pacific, is a mere 70 miles, but this volume would increase train scale, eliminate truck congestion in the port, and allow draymen to work much more productively. Because network hubs are not intended to support a specific, adjacent hinterland, they offer the flexibility to accommodate unanticipated markets that may develop in the future. These may be single or multiple railroad routes.
New solutions for the future
In the North American intermodal system, conflicting economic and operational interests have created imperatives to expand both scale and scope, a situation that is creating diseconomies. As container ships get bigger and carry larger numbers of boxes, on-dock space increasingly is needed for vessel operation (diseconomy of scale). Moreover, on-dock rail yards originally were designed for loading multiple containers to single destination terminals, but meeting demand today requires loading to more, not fewer, intermodal terminals, in order to realize drayage savings (diseconomy of scope). In short, ocean and rail carriers are pushing for scale, which is a barrier to an effective drayage system, which demands scope. All of this is working against a meaningful intermodal product that benefits all of the providers as well as the exporters, importers, and domestic shippers that are the ultimate customers of the intermodal system.
This must change. The current intermodal system may become incapable of handling additional traffic because everyone is fighting the last war, not the war to come. There is no time to lose. What will happen if the Trans-Pacific Partnership (TTP) treaty is approved, and that causes a big leap in trade volumes? Consider also the effects of the recent West Coast port congestion and labor actions. When the freight transportation system is constrained, it has a measurable impact on the North American economy.
For those reasons and more, current capacity problems demand new solutions to ensure that the North American intermodal system continues to support domestic and international freight mobility in a reliable, economic, and environmentally benign manner. Introduction of network hubs promises to do just that.
1. This article will refer to "intermodal operators" to encompass both ocean carriers and domestic intermodal shippers.
2. The connectivity method is frequently the result of intense negotiations between the railroads. Critical mass is required for steel-wheel interchanges; however, other factors can obviate that solution. For example: terminal capacity at the first railroad's origin or destination intermodal terminal, or at the second railroad's origin intermodal terminal; train interchanges between the two railroads; interchange location; or the involvement of more than two railroads.
3. CREATE was intended to be a public-private partnership that would improve railroad-to-railroad connectivity and fluidity within Chicago. Other than a handful of projects, it has failed to deliver any significant improvements.
4. Since NWO is for containers only, trailers must still be interchanged through Chicago.
5. Increased volume could lead to multiple trains a day, which would improve the intermodal product by reducing train headway (the distance or time between vehicles in a transportation system).