Rover Retrospective – A Case Study of Project Risks

Industry participants know large infrastructure projects are risky. From origination to operation, risks and uncertainties regarding feasibility, costs, and schedules are myriad and difficult to quantify. The stakeholders are many and often have opposing interests. The energy value chain, federal and state regulators, capital markets, private landowners, special interest NGOs, and the energy-consuming public all contribute to a project’s success or failure. Despite all of this, critical infrastructure is built, and America’s lights stay on.
When things go right, there is recency bias, where we underweight historical trends. When things go wrong, 20/20 hindsight is deployed and questions are asked: What did the company know and when did it know it? Could more of the risks have been anticipated and mitigated? The Rover Pipeline effectively realized 3 Bcf/d of its full service capacity (3.25 Bcf/d of natural gas) this August, achieving its stated goal of transporting natural gas from the Marcellus and Utica Shale basins across Pennsylvania, West Virginia and Ohio, and then north into Michigan and Canada. It wasn’t easy. A look at the Rover Pipeline project provides an illustrative case study for identifying the risks inherent in the planning and execution of large energy infrastructure projects – in this case, an interstate natural gas pipeline system.
As we learned, Rover was at all times optimistic in its ability to bring the project into service quickly, but our customers, whether developers, investors, shippers or commodity traders, were able to gauge just how optimistic Rover was being, and were consequently able to mitigate the risks that came from overexuberance.

Along the way, the majority owner of the pipeline, Energy Transfer Partners, lost a lot of market value and credibility, and other stakeholders lost a lot of money, as many of the inherent risks of projects were realized - causing delay after delay. Were these delays tolerable when viewed against the project’s long term return on investment?
On average, according to LawIQ’s platform, from filing of the Certificate application, pipeline projects estimated to cost $1 billion or more take 23 months to obtain a FERC Certificate, and 34 months to be fully placed in service. Energy Transfer planned for Rover to receive its FERC Certificate in nine months and be placed in service in 21 months. Ultimately, Rover took 50 months to complete construction of its 711 miles of supply laterals, two 42-inch pipes, and related compression and metering facilities. The original estimated cost was $4.216 billion, with $1.762 billion allocated for labor. In its initial annual report, Rover reported costs of $5.130 billion, and we expect final costs could be closer to $6 billion. So, what happened?

The Plan was Aggressive

In its original Certificate application filed in February of 2015, Rover had requested FERC approval by November of that year, and projected putting Phase One of the project in service in December of the following year, and Phase Two by June of 2017. Rover’s schedule set an expectation upfront that it would receive its Certificate 14 months ahead of the average for similar projects and would be placed in service 13 months ahead of the average.
A stated reason for being “aware that it is proposing an ambitious schedule” was to honor shipper commitments and market demand. It’s likely competitive dynamics also influenced Energy Transfer setting these expectations, as Enbridge and DTE Energy’s Nexus project was at the time also being contemplated and similarly positioned.

Environmental Review Phase – Unexpected Permitting Challenges

By November of 2015, FERC had not yet issued the Environmental Impact Statement (EIS) (the first of the major post application milestones), prompting a strongly worded letter from Rover suggesting that FERC was taking too long, particularly since the project had pre-filed. A final EIS was issued on July 29, 2016.

In August 2016, we summarized in Rumor Has It – Rover Pipeline – Past, Present and Future, the revisions being made by company officials to Rover’s forecasts for receiving its Certificate and getting Phase One of the project into service – Q4 2016 and November 2017, respectively. To meet these new dates, representing a full year delay from the forecasted dates in the application filing, the project would still need to receive 16 major federal permits, including some delegated to various state agencies, such as those for water quality certifications. Water quality permits would become an issue, but first came the bats.

In the category of unexpected risk, a hurdle Rover encountered between EIS and approval related to a cooperating agency permit needed from the U.S. Fish and Wildlife Service (FWS) for the mitigation of impact to a certain species of bats located along the proposed path of construction. We discussed this in Caution. Low Flying Bats - Unlikely Project Risks. The FWS rejected the initial Myotid Bat Conservation Plan, causing another unexpected delay, and a condition was added to the final EIS requiring all tree clearing activities be completed by March 31, 2017 – the start of which required a Certificate.

By Q4 of 2016, the pressure was really on to get FERC approval. Our dynamic forecast (shown in the chart below) estimated approval was approximately 100 days away. It turned out to be 150 days, arriving on February 2, 2017, and just two days before FERC Chairman Norman Bay was to step aside and leave the Commission without a quorum. This was a rare break for the project.

The next chart shows our heuristic estimate. It is based on statistical inputs at the start of a project, but is not updated dynamically over time. When a milestone is reached, the timeline is updated to reflect the actual time elapsed. In this case, the actual time required from pre-filing to a FERC approval was 38.5 months. And, our estimate for the time it would take for construction was 16.5 months. Adding the actual times to the remaining estimate yielded an In Service Date (ISD) in June of 2018. On ETP’s Q4 2016 earnings call, the company estimated an ISD for Phase One of July 2017.

Let Construction Begin

Those in the industry and interested investors know how this story goes – more unexpected delays in the construction phase, the difference being many were avoidable. Lawsuits to secure eminent domain land rights, a drilling fluid spill while conducting horizontal directional drilling and a subsequent “stop work” order from FERC, “cease and desist” orders for failure to meet water quality standards in Ohio and then West Virginia, were just some of the issues encountered during Rover construction. And the costs of these delays were not limited to ETP.
Notably, commodities traders took large losses (over $100 million) on positions that relied on the timely completion of pipelines, including Rover, transporting gas to market from the Marcellus/Utica basins, and exploration and production companies had to deploy hedging strategies to protect against prolonged basis weakness caused by pipeline project delays.

Can Future Rovers be Anticipated or Mitigated?

So what should infrastructure developers, shippers and commodities dealers do to anticipate Rover-like problems in future cases? First, by using LawIQ’s platform that provides objective projections for key milestones based on historical data, everyone involved would be able to tell just how aggressive a proposed schedule is. In the coming weeks, we will be rolling out similar information about the costs of comparable projects, which will allow our clients to determine how aggressive the cost projections are. Finally, once problems start cropping up, calls to LawIQ’s experts, like the hundreds we fielded on Rover, can help our customers scope out the risk to the project timeline based on our review of the underlying documents and our expert knowledge of the applicable legal and regulatory issues.

Insights Coming Soon