Permian is Both Overbuilt and Underbuilt Depending on the Product

In 2018, 2019 and 2020, Arbo issued a monthly report in which we monitored the buildout of pipelines designed to transport various products from the Permian Basin to markets. Those reports stopped at the end of 2020 as the last two pipelines reported on, Permian Highway and Wink to Webster, were under construction. That last buildout was occurring just as the markets for oil and gas were collapsing from demand destruction caused by the pandemic, but now U.S. producers are being asked to increase production to replace Russian oil following the invasion of Ukraine. The interesting factor is that most agree there is plenty of pipeline capacity for crude and NGLs from the Permian, but what may be in short supply is takeaway capacity for the associated gas that will likely be produced as crude production ramps back up. Many market watchers believed that Energy Transfer leapfrogged the competition when it announced it was exploring a Permian expansion that would include the use of some of its existing Texas assets.

We think it is too early to declare Energy Transfer the odds on favorite in the race to provide additional takeaway capacity. The pipeline it has publicly mentioned would still require the construction of 260 miles of new pipeline to connect Midland to existing Energy Transfer pipelines near Fort Worth. Based on the last buildout in Texas, a pipeline of that length has no time advantage over a longer pipeline. However, from a timing perspective, converting one of the recently completed liquids pipelines may be the fastest way to bring additional gas capacity online.

Energy Transfer’s Proposed Pipeline

On its earnings call in February, Energy Transfer noted that “to address the growing need for additional natural gas takeaway from the Permian Basin, [it is] diligently evaluating a takeaway project that would utilize existing Energy Transfer assets along with new build pipeline providing producers with firm capacity to the premier markets of Katy, Carthage, Gillis and Henry Hubs. This pipeline project would include the construction of a new approximately 260-mile pipeline from the Midland Basin to our existing 36-inch pipeline southwest of Fort Worth, parallelling existing rights of way.” It is interesting to note that the company indicates that the 260 miles of new pipeline will parallel existing rights of way, but not necessarily their own.

Many observers seemed to be convinced by the company’s observation that it could complete its proposed project “much more quickly than our competitors' options at significantly less cost by following an existing right of way along the majority of the route.” While the use of existing assets to connect the new pipeline to premier markets would likely lower the cost, as we discuss below, the company’s belief that it could “complete construction of the project in 2 years or less once we have reached FID” is not a decidedly significant timing advantage. Also, the company’s asset map does not include a company-owned pipeline that would parallel the entire 260 mile route.

Longer Route Does Not Mean Longer Time to Build

The mere fact that the Energy Transfer proposal involves the use of some of its existing pipelines does not necessarily mean that it has a timing advantage over other pipeline proposals because it still requires the construction of 260 miles of new pipeline. Whether a company is building 50 miles of pipeline or 500 miles of pipeline, the time it takes to plan and build the pipeline is not heavily dependent on the length of the proposed pipeline. That is because most activities required to complete the pipeline from land acquisition all the way through construction can run in parallel for each of the segments of pipeline. Therefore, the time it takes to build is really dependent on how many teams of people a company assigns to a project. As many in the industry have observed, a company does not build a 250-mile pipeline or a 500-mile pipeline, it really builds five or ten fifty-mile pipelines simultaneously.

During the last buildout of Permian pipelines, we monitored the process for the various pipelines from the commencement of land acquisition through to in-service.

As seen above, the typical time period to begin land acquisition is about 24 months before the projected in-service date and that time period didn’t vary widely for short or long pipelines. Thus, even Energy Transfer’s statement is consistent with a two-year cycle from the commencement of land acquisition, which usually commences following FID, to being in-service.

Characteristics That Could Shorten Timelines

One key way to shorten the timeline from planning to in-service is to convert an existing pipeline from use for liquids to gas. Looking at four FERC projects that involved the conversion of a liquids pipeline to use for gas service, the time from approval to begin the conversion to a commencement of at least partial in-service shows that the time to convert can be very short.

As seen above, KM KN Interstate Pony Express Pipeline was able to begin partial service within about six weeks after it received approval to convert the 850-mile line in that case. That project went into full service about three months after receiving approval to commence the conversion, as did El Paso for its conversion of 94 miles of Line 1904. Those times are much shorter than the time required for construction of even a 50-mile pipeline.

During the last Permian buildout, it appears that the liquids and crude pipelines may have actually overshot the need for additional pipeline capacity.

As seen above, the data from Arbo’s Liquids Commerce Platform shows that Gray Oak and Cactus II have been reducing their tariff rates in an effort to encourage shippers to use available capacity. There may be others as well that are not very full and could convert the pipeline to gas use as either an interim or permanent solution to the apparent need for additional gas capacity from the region.

Easements as an Early Indicator

As the visual above showed, a pipeline’s effort to begin accumulating the easements necessary to construct a planned pipeline typically starts about two years before the pipeline is put into service. So far, we are not seeing any efforts by Energy Transfer to start that process, which may indicate that any future capacity is at least two years away. However, if Energy Transfer already has many agreements that allow it to lay a second pipeline in the same easement, it would not necessarily need to obtain new easements in the portion of the path that is co-located with their existing pipelines. However, as we noted above, Energy Transfer does not appear to have any contiguous path between Midland and Fort Worth, which means at least some land acquisition will be needed.

For pipelines that are looking to convert from liquids to gas use, the easements they obtained may make it easier for some to convert than others. For example, it appears that the typical easement form used by the Gray Oak pipeline provided that the pipeline to be placed in the easement could only be used for “the purpose of transporting crude oil,” but not expressly for natural gas. Conversely, the typical easement form used by the Epic Y Grade line allowed for the covered pipelines to be “used for the transportation of natural gas, gaseous products, crude oil, and other hydrocarbons or minerals, whether in gaseous or liquid form.” Clearly, whether a pipeline can be easily converted to use for gas will be heavily dependent upon such issues, but there may be a pipeline with excess capacity that already has the right to convert to gas use and could much more quickly provide such service than constructing a new pipeline. So for now, we do not see the Energy Transfer project as having a significant advantage over any other greenfield pipeline project. However, a project that could utilize an existing liquids pipeline just may have such an advantage.