Pipe, Rail or Truck, Comparing the Ways to Move a Barrel of Oil Safely

Here is a fact that will not fit on a protest sign: when you block a pipeline, the oil does not stay in the ground. It goes onto a train. Or onto a truck. Or onto a barge. The barrels still move, because people still need what is in them, and the only question is whether they move through a steel tube buried in the earth or down a rail line that runs through the middle of somebody's town.

That is the whole argument of this post, and I am going to make it carefully, because the other side has a case and because the stakes include human lives. Folks, if you care about spills, if you care about safety, if you care about the water, then you ought to care about which way the oil travels. The people who fight pipelines say they care. Let us see what the comparison shows.

Defining the Question

The question is not whether oil should move. It does. Refineries need crude, and the country burns, refines and manufactures with petroleum every day. The question is which mode of transport, for a given job, produces the least harm per barrel delivered.

To answer that you need a measure. The honest measure is incidents, spills and fatalities per unit of oil moved over a distance, not the raw total for each mode, because pipelines move an enormous volume and a raw count can mislead in either direction. Safety regulators in the United States keep records of this. The Pipeline and Hazardous Materials Safety Administration, called PHMSA and part of the Department of Transportation, collects reports on pipeline incidents, and the Federal Railroad Administration and PHMSA together track hazardous-materials incidents on the rails. I will not quote a statistic I cannot check in front of you. Instead I will tell you what the kinds of numbers show and where to go find them.

The Comparison Table

Here is the comparison in plain terms. The entries are general characteristics, and I encourage you to compare them against the current PHMSA data tables yourself.

A word of fairness about that table. A pipeline spill, when it happens, can be a big one, because a pipeline holds a lot of oil and may leak before a sensor catches it. A leak into a river is a serious event that deserves a serious response. I do not minimise it. The size of the gap is disputed: studies from the Fraser Institute put the pipeline incident rate well below rail's, while the Railway Association of Canada argues that the two are about equally safe once only crude is counted. But the weight of the evidence I have seen is that a barrel on a pipeline is less likely to be spilled than a barrel on a railcar or a truck, and the consequences of the others show up as human casualties, not just cleanup bills.

The Case That Changed the Argument: Lac-Mégantic

On July 6, 2013, the place was Lac-Mégantic, a small town in Quebec. A runaway train carrying crude oil from the Bakken formation in North Dakota rolled down a grade into the middle of town and derailed. The oil caught fire and exploded. Forty-seven people died, and much of the downtown was destroyed.

I raise this not to be gruesome but because it is the best-documented example of what oil on rails can do. The train was a Canadian operation on a Canadian line, and its crude had come from American oil fields. Its tank cars, an older design known as the DOT-111, were widely criticised for puncturing and rupturing in derailments. The disaster set off a wave of rule changes on both sides of the border, including tougher tank car standards and phased retirements of the older designs, and a revived debate over how much crude should ride on rails at all.

It is no insult to anyone who works the railroads to say this. Railroads are essential, and for many freight jobs they are the safest and most efficient option we have. But a mode that is excellent for grain and automobiles is not necessarily the best for a million barrels a day of flammable crude. And the fight against pipelines pushed more crude onto rails in exactly the years when the Bakken was booming and pipeline capacity was thin. Crude by rail climbed to more than a million barrels a day on average in 2014, a figure you can check in the Energy Information Administration's data, and the reason was not that railroads were the first choice. It was that the pipelines were not there.

The Dakota Access Fight

Consider the Dakota Access Pipeline. It carries crude from the Bakken fields in North Dakota to a hub in Illinois, a route of about 1,170 miles. The project drew a long protest in 2016 centred on its crossing of the Missouri River near Lake Oahe, upstream of the Standing Rock Sioux Reservation. The tribe and its supporters raised concerns about water and sacred sites, and those concerns were not frivolous. A government that crosses a river near tribal land owes the tribe a serious hearing, and the consultation process was a legitimate subject of litigation, which went through the federal courts for years.

Now put the safety question beside the politics. Before the pipeline was running, a great deal of Bakken crude moved by rail. Once the pipeline opened in 2017, it took on a large part of that flow. The point is not that the pipeline can never leak. The point is that opposing it did not make the oil vanish. It decided which route the oil would travel, and the alternative was more unit trains through more communities.

Notice that a sincere opponent can hold the water concern and still face the comparison. If the underlying fear is a spill into a river, then a train crossing a bridge over the same river is not a safe alternative. A derailment on a river bridge is the same fear, in a more flammable package.

The Line 5 Dispute in Michigan

The Enbridge Line 5 dispute raises the same question in a different place. Line 5 carries light crude and natural gas liquids from Wisconsin through Michigan to Ontario, and a section of it crosses the Straits of Mackinac, the narrow waterway where Lake Michigan meets Lake Huron, on twin pipes laid on the lakebed in 1953. The lines are old, and the Straits are a place where nobody wants a spill. An anchor strike in 2018 dented the line and cut nearby electric cables, which showed how real the worry is.

The State of Michigan has fought in court to shut the line down, and Enbridge has proposed building a tunnel under the lakebed to house a new pipe. I do not pretend this is a simple case. The worry about the Straits is legitimate, and a tunnel is a serious engineering response to it.

But here is the question the opponents should answer. If Line 5 is shut, where do the barrels of product the line carries each day go? The line feeds refineries in the region, and a good part of the propane for the Upper Peninsula comes through it. The replacement would be trucks and railcars, and the Upper Midwest has icy winters and two-lane roads. A truck-borne replacement means thousands more trips on those roads, with the crash risk that comes with them. I will not quote a daily capacity figure here, and you should confirm the current number with the company and with the state's own filings. The principle does not depend on the digit. A closure moves the risk, and the new risk may be bigger.

Which Mode Suits Which Job

Let me be fair to every mode, because the right policy is not "pipelines good, everything else bad." Each has a place.

• Pipelines are best for steady, high-volume, long-distance movement between fixed points, such as a field to a refinery. They are the cheapest and, in general, the safest by the measures that count.
• Rail is best when the route needs flexibility, when volumes are modest or seasonal, or when the destination changes. It also serves places that no pipeline reaches. With modern tank cars and careful routing, rail has a role.
• Trucks are best for the last miles: delivering from a terminal to a gas station, a farm or a job site. Nobody proposes running a pipe to every driveway.
• Barges and tankers move enormous volumes cheaply on water, with their own record and their own regulation.

The mistake is to treat the choice as moral rather than practical. A pipeline is not a sin and a railcar is not a virtue. They are tools, and a sensible policy fits the tool to the job and holds each to a high safety standard.

The Strongest Objection, Stated Fairly

The best argument against building more pipelines is this: a pipeline locks in decades of oil consumption, and every new line is a commitment to keep burning fossil fuels. If you want to reduce emissions, you should not build the infrastructure that makes oil cheap and easy to move. Better to let the oil be expensive and scarce, so that demand falls.

Take that seriously, and then answer it in order.

1. The demand is still there. Oil use is driven by what people drive, fly, heat and manufacture, not by the pipeline map. Unless demand actually falls, blocking a line changes the route, not the quantity. The barrels move by train and truck instead.
2. The substitute emits more. Moving oil by rail or truck burns diesel and produces more greenhouse gas per barrel than a pipeline does. If emissions are your concern, the policy you chose adds to them.
3. Scarcity falls on the poor. Making fuel more costly by constraining transport does not hurt the executive with a hybrid. It hurts the family on a fixed income with a long commute.
4. The costs of safety are hidden. The extra risk to a town that hosts a rail line, or a road that hosts a tanker, is borne by people who never got a vote on the pipeline fight.

The cleanest version of the climate argument is to reduce demand honestly and openly, by persuading people and by building alternatives that really compete. Restricting the safest transport mode in order to make oil painful is a way of getting the result without making the argument, and it fails at its own goal.

The Constitutional Frame

Pipelines that cross state lines are a classic subject of federal authority under Article I, Section 8, Clause 3, which gives Congress the power "To regulate Commerce with foreign Nations, and among the several States." The framers wanted a single national market, free of the trade barriers that had plagued the Confederation, and an interstate pipeline is a modern artery of exactly that market. The Keystone XL project showed the other side of the coin: a pipeline that crosses an international border needs a presidential permit, and the Obama administration denied it in 2015, the Trump administration issued a permit in March 2017, and the Biden administration revoked the permit on its first day in 2021. A project that rises and falls with the occupant of the White House is not a project anyone will risk billions on.

That is a failure of the rule of law in the plain sense: the rule is whatever the current officeholder prefers. A republic works better when permits are issued under clear statutory standards, reviewed on a predictable schedule and decided on the merits. For how that kind of delay has affected American building more broadly, see We Raised the Empire State Building in About a Year, So Why Not a Power Line. And for the larger picture of what America's energy sector has accomplished when it was allowed to build, see How America Went From Importing Natural Gas to Shipping It Worldwide.

What a Safe System Would Do

If we wanted safety above slogans, the policy would look like this.

• Hold every mode to measurable standards and publish the incident data per barrel moved, so the public can compare.
• Speed up decisions on pipeline permits, with clear deadlines and judicial review on a short clock, so that safer options are not delayed into extinction.
• Invest in inspection and leak detection, including internal inspection tools and modern sensors, for the older lines already in the ground.
• Retire the worst equipment on every mode, as the tank-car rules did after 2013.
• Be honest about the tradeoff. Anyone proposing to close a line should be required to say where the oil goes instead.

You can read about the way government rules shape what energy costs in Four Pieces of Every Gallon and Why the President Controls the Fewest, and the broader record of the nation's emergency stockpile in From the 1973 Embargo to the 2022 Sale, a Strategic Petroleum Reserve Timeline.

The Case in Brief

1. Oil moves by pipeline, rail or truck, and blocking one mode sends the barrels to another.
2. Incident studies generally find pipelines spill and kill less per barrel moved than rail or trucks, though the size of the gap is disputed and pipeline spills can be large.
3. The 2013 Lac-Mégantic derailment, which killed 47 people, shows the human cost when crude rides the rails through towns.
4. The Dakota Access and Line 5 fights are legitimate disputes about water, but closing a line shifts the oil onto less safe routes rather than removing it.
5. The right policy fits each mode to its job, speeds permitting, strengthens inspection and publishes the numbers.

What you can do: Search the PHMSA incident data tables online, compare pipeline and rail hazardous-liquid incidents for yourself, and bring the numbers to the next public hearing on a pipeline or rail route near you.