crude oilLast month's publication of the State Department's latest environmental impact report on the Keystone XL pipeline project has sparked great interest in the logistics of shipping crude oil by rail. As described in a long article in the Washington Post, the availability of a rail option for oil sands crude could prove to be a crucial element in determining whether the pending decision to permit the pipeline to cross the US border would actually affect Canada's oil sands output, and thus its greenhouse gas emissions. As the article makes clear, however, oil's rail trend is already well underway , thanks to the surge of "tight oil" production from shale formations. Moving crude oil by train is experiencing a "Back to the Future" moment.

Oil shipments in rail cars are nothing new; the practice dates back to the earliest days of the oil industry. In fact, control of key railroad routes for oil and petroleum products was an important aspect of the US government's anti-trust case against the original Standard Oil a century ago. My first exposure to crude-by-rail was in the 1980s, when significant quantities of heavy crude from California's San Joaquin valley were routinely transported to Los Angeles refineries by dedicated "unit trains", because there wasn't sufficient pipeline capacity available.

The same dynamic applies today, with the rapid expansion of tight oil production in North Dakota's Bakken fields quickly outstripping the capacity of the state's few existing pipelines to transport the oil to market. A tank car loading rack requires much less time and money to build than a new pipeline or pipeline expansion. US railroads are also eager for the traffic, since coal deliveries, which accounted for 45% of US rail traffic in 2011, fell by nearly 11% last year as natural gas eroded coal's share of power generation. Meanwhile oil shipments by rail grew by 46% in 2012.

Precise data on just how much crude oil is currently moving by rail are hard to find. The American Association of Railroads doesn't differentiate between crude oil and refined petroleum products, which until recently accounted for most oil-related rail shipments. The US Energy Information Agency (EIA) reported last summer that crude oil had grown to roughly 30% of total petroleum rail deliveries, which would equate to around 300,000 barrels per day (bpd) on average for 2012. Yet EIA's analysis of recent trends suggested that crude-by-rail increased by nearly 250,000 bpd last year alone. The CEO of the Burlington Northern Santa Fe recently indicated that his railroad's total oil-related shipments alone could expand to around 1 million bpd, roughly double today's level.

It would be easy to conclude that all this growth reflects a temporary expedient, until North American pipeline capacity can be expanded and realigned to match rising output and the reversal of long-standing import trends. That view is clearly not shared by oil companies and traders who are lining up to purchase or lease new tank cars for this service. Perhaps that's because rail provides a degree of flexibility that would be nearly impossible to match by pipeline. For example, it creates an opportunity to supply domestic crude to East Coast refineries like Delta Airlines' Trainer, Pennsylvania facility, which had previously become uneconomical to operate on a diet of imported crude cargoes. Similarly, even if a pipeline from North Dakota to the San Francisco Bay Area could be justified economically, it would likely never receive the necessary permits. Yet Valero's Benicia refinery might soon receive up to 70,000 barrels per day of Bakken crude by rail.

Railroads are also surprisingly efficient. At an industry average of 480 ton-miles per gallon, my analysis indicates that shipping a barrel of crude from North Dakota to a refinery in either Houston or Philadelphia consumes a quantity of diesel fuel equivalent to just 1% of the energy content of the oil, while adding slightly over 1% to the typical well-to-wheels emissions for gasoline refined from it. That's higher than for pipelines, but not by enough to render the option unattractive.

Pipelines remain the preferred option for moving high volumes of oil safely over long distances and, when capacity exists, are usually cheaper for shippers. However, rapidly shifting sources of production and the high capital costs of new pipelines, combined with an increasingly challenging regulatory environment, could provide a durable opportunity for oil-by-rail, just as it has for moving petroleum products and ethanol by train

A slightly different version of this posting was previously published on the website of Pacific Energy Development Corporation.