No one could be blamed for having the wrong impression of pipelines. Views tend to align with politics, and media coverage often casts these silent transport methods in one extreme light or another. The facts about pipelines are quite revealing, but there is a lot of data to discuss before giving a final safety rating.

To begin with, pipelines are ubiquitous. Not only do pipelines –a broad term here to include transmission, gathering, and distribution lines– carry much of the raw material needed to power the U.S., but they are virtually everywhere. According to the U.S. Department of Transportation, there are over 2.8 million miles of pipeline running across the U.S. alone. That distance could run back and forth between San Fransisco and New York City over 1,100 times.

Rather than repeating the Golden-Empire route for a thousand lengths, the pipeline network in the U.S. is embedded in each and every state. It has been described by former secretary of transportation, Norman Mineta, as arteries transporting the critical lifeblood of the American economy. Like the human body, America requires energy to be efficiently delivered to every last region.

The first pipelines emerged in the U.S. in the 1860s. The innovative leap from the first rudimentary troughs to sophisticated pumped pipelines took only two years. Over 150 years on, pipelines are state of the art and moving impressive volumes on a daily basis. Today, pipeline infrastructure moves nearly 30 trillion cubic feet of natural gas and around 16 billion barrels of crude oil and refined products each year.


Safety Record

Pipelines operate day in and day out. They move so much volume that it is difficult to conceptualize. That high volume, though, means that when rare incidents do occur, they often result in large losses. In fact, if only one percent of oil product transported in a year spilled, that would leave the nation coated in some 160,000,000 barrels of oil.

That frightening look at volumes can lead to serious questions about whether we can view pipelines as safe, because even a near-perfect record could mean enormous damage when the rare problem occurs. Fortunately, pipelines do far better than losing one percent. The available data from industry and validated by the federal government is that 99.999 percent of the product that enters a pipeline arrives at its end point. This number could even be as high as 99.999997 percent, but we will use the more modest statistic.

A skeptical observer would note that this figure comes from the industry itself. But then, who would have better data than the industry, and who would know the exact amount better than pipeline operators and those paying for the product? At the end of the day, they have an interest in the most product making it to the other end as an economic matter, as well as a reputational one. Moreover, the numbers are not seriously debated. Environmentalists are content to accept as true yet harshly criticize the 0.001 percent of losses as catastrophic. Using the 16 billion barrels of crude and refined oil products, this lost product rate would mean roughly 160,000 barrels erroneously outside of pipelines in a year. Of note, natural gas leaks or spills are less likely to occur than oil.

Any analysis would be remiss without mentioning pipeline technology for remote monitoring and shutoff capability. While some may call to mind hazardous liquid rushing through basic steel pipes, the truth is that pipelines are advanced transportation methods that include innovative technology to determine volume and flow capacity, pressure, and more to detect and resolve leaks in real time. Modern pipeline operators do not need to wait until someone stumbles across an oil-coated field to know there is a problem. With pipelines being developed today, the operator may know and be able to remotely shut down an interval of a pipeline as soon as an issue arises.


So what explains the 0.001 percent leakage?

Pipelines tend to operate safely when not impeded or damaged. While sometimes pipelines do become over-pressurized, corrode, or contain a critical flaw, these events are rare. In 2020, across all pipelines in the U.S., there were 581 total incidents. Fewer than half were considered significant. Of those damage incidents, 10 percent can be attributed to natural forces such as weather and earthquakes; intentional acts of destruction or terrorism; or accidents like vehicle wrecks into critical infrastructure.

Then, there are professional third-party accidents. At least 12 percent of the leaks and spills that do occur do not result from any negligence or flaw in the pipeline itself or its operation, but on the part of excavators digging into the earth without knowing what is below. Fortunately, there is an entire industry focused on damage prevention.

Looking at the 99.999 percent safety record and knowing external causes explain the few hundred rare incidents across the 2.8-million-mile network annually, it would be reasonable to conclude pipelines themselves and the operation of them is safe. It still leaves us with a lot of lost hazardous product, based on volumes. The percentages tell us that for every barrel of oil transported, 0.05 ounces may be lost. Or in larger terms, that when 100,000 barrels are transported, one whole barrel would be lost. Of course this represents an average – every barrel does not seep 30 droplets of oil at the seams – the norm is that no product is lost. Data suggests that for 73 percent of leaks, the product lost is around 6 barrels or less. For over half of U.S. incidents in 2020, fewer than 5 barrels of highly volatile liquid was released, or fewer than 50 barrels of other liquids.

There is also an argument that the 99.999 percent figure should not be taken to be a prediction for future leaks. Year over year, data suggests pipelines are becoming safer and losing less product. In the last five years, even with 10 percent more volume moved and pipeline mileage added, leak incidents fell approximately 20 percent. Still, when spills or leaks do occur, they can be huge and devastating to public safety, ecology, and more.

When evaluating safety, we must recognize that harm does occur. If the 0.001 percent leak rate, or 99.999 percent effective delivery rate, still sounds too unsafe because of high volumes, we should look to the vulnerable areas possibly impacted. If leaks or spills are happening there, then the fraction of a percent of total leak will matter more to the final safety rating.


What about water

Waterways and wetland areas are ecologically diverse and may host sensitive plants and animals. Environmental activists routinely point to water crossings as grounds to block or cancel proposed pipelines. Yet, the data on pipelines near water does not suggest they are causing havoc on the water.

In 2013, PHMSA concluded that there were 18,136 inland water crossings in the United States. The same study found that of all 5,094 incidents over the 21 years from 1991 to 2012, only 0.3 percent (or 20 total incidents) occurred at inland water crossings. Other studies show a higher percentage of incidents near water crossings. In 2016, PHMSA found that across six years from 2010 to 2016, one percent of pipeline leaks impacted or were even near a body of water. The second study included a broader definition for what was considered near water, and even then revealed a single percent. Statistically speaking, pipelines are not a threat to watery environments.

Beyond water crossings, the question of where spills and leaks occur is interesting. According to research done in the U.S. and Canada, very little hazardous material actually leaks or spills in the line pipe between facilities.

During the period 2003 to 2013, only 17 percent of occurrences took place within the actual line pipe. The majority of pipeline accidents occurred at facilities, which include, for example, compressor stations, gas processing plants, pump stations, terminals, transmission line pig traps, etc. Spills that occur in these areas are often contained within the facility, which may have secondary containment mechanisms and procedures.

With this added context, the question of pipeline safety narrows. The percentages indicate that the overwhelming majority of hazardous product arrives at its end point, and very minimal percentages leak at all or in sensitive areas. Further, of the leaks and spills that do occur, many occur at natural joints in the infrastructure network.


Time for a safety verdict?

Pipelines are constantly moving enormous volumes underneath our feet. They do so silently under roads and bridges, under fields and streams, and under our buildings and homes. We know that they carry their payload from end to end with well over 99 percent effectiveness, and that the leaks that do occur are not in ecologically sensitive areas. We also know that technology continues to advance, such that leaks can be detected remotely and the pipeline shut off within moments to avoid adverse environmental impacts.

Nonetheless, the huge volume moved means that the fraction of a percent lost can still be a huge volume of dangerous material. This must be factored in. It is difficult to say that a transportation method with an effective delivery rate so close to 100 percent is unsafe, but the full context may lead rational people to conclude its flaws are grave. This is where many would point out that where pipelines are concerned, given how much we need the resources they carry, we cannot let the perfect be the enemy of the good. And 99.999 percent is pretty good.


What are the alternatives

If you do not arrive at an objective view that pipelines are safe in isolation, perhaps consider them in a wider context. If hazardous and volatile material like oil and natural gas must be transported –and they must– then the next methods available are railcars, tanker trucks, and marine vessels.

These options also boast impressive safety records to be sure. In most cases, because of innovation, technology, high volumes, and regulation, these transportation methods also transport with nearly 99 percent effectiveness. There are still major factors to consider.

Efficiency and harm are the the two categories most suited for analysis. But within these, there are individual factors like cost to transport, speed of transport, product arrival effectiveness, dollars of damage, human injury or ecological harm, environmental emissions to operate the transportation method, and the risk to or from third party actors. Pipelines do not come out in first place for each of these analyses, but very real drawbacks impact the alternatives in ways that reasonably put pipelines in the lead, or at least contextualize that the drawbacks are a feature of moving hazardous materials, not bugs specific to pipelines.

First is that each of these methods produces its own emissions along the way. Engines burning coal, gasoline, and oil power these vehicles as they carry enormous payloads. This is an environmental concern more than safety, but with pipelines, the two are one and the same. Next is that these operate in unpredictable and dynamic settings. On a highway, a tanker truck is not only subject to the driver and engineering standards of the vehicle, but weather phenomena, driving conditions, and other drivers’ behavior. Rail similarly encounters track deficiencies and crossings where drivers and other obstacles interfere with the smooth transportation of hazardous materials. On the seas, movement of hazardous material occurs near the water continuously, which has to be loaded and offloaded from the vessel, and the ship then encounters waves and variability on the water.

By contrast, pipelines are almost exclusively underground and insulated from weather, vehicles, and living things. They do face the risk of excavation damage, but following the call-before-you-dig laws considerably lessens that risk.

When all is said and done, settling the question of pipeline safety requires looking at the numbers and comparing other modes of transportation. Pipelines are already doing the work of moving mind boggling volumes of oil and natural gas under our feet today. The safety record, in context, leaves us to conclude that pipelines are safe.


Written by Benjamin Dierker, Director of Public Policy


The Alliance for Innovation and Infrastructure (Aii) is an independent, national research and educational organization. An innovative think tank, Aii explores the intersection of economics, law, and public policy in the areas of climate, damage prevention, energy, infrastructure, innovation, technology, and transportation.