But I Would Pipe Five Hundred Miles - Evaluating Economics of a New Natural Gas Pipeline (Part 4)

It’s no secret that a long list of pipeline projects have been proposed to help move natural gas out of the Northeast production areas. But if you were a Marcellus or Utica producer, how would you decide whether you were interested in new capacity that hadn’t been proposed or built yet? Of course, pipeline companies have armies of engineers, cost estimators, and market analysts to bring one of these monster projects to fruition. But for anyone else, particularly in the early stages, how do you even know it’s a reasonable idea? For anyone testing a concept, you need a way to ballpark some scenarios for a new pipe. We’ve been running a blog series on our RBN Pipeline Economics Estimation Model, a quick, rule-of-thumb “sanity test” for new capacity. Today, we wrap up our walk-through of the model, with a real-world example to gauge the accuracy of the model, and then with a discussion on how the model can be used to measure economies of scale in picking the minimum volume you probably need for a new pipeline.

As we discussed in Part 1, there are several factors a potential shipper needs to assess in deciding whether to pursue—and, ultimately, to commit—to a new pipeline project. These factors can be grouped in two big categories—how much is it worth to move gas from Point A to Point B (i.e. the price differential between your wellhead price and the price at the delivery point), and what will it cost to get there? But to estimate what that cost might be, you have to first know how much the pipe will cost to build, and to know that you have to decide how large the pipe needs to be. In Part 2, we walked through an example of how to go about figuring out the pipe size using our back-of-the-envelope Pipeline Economics Estimation Model. Our model would probably make a pipeline engineer laugh, but it gets you close. In our example, you’re a Marcellus producer who sees an opportunity — a $4.00/MMBtu price differential (on average for the year) between the Tennessee Gas Pipeline (TGP) Zone 4 pricing hub near your supply area and the Boston market — and you want to know if it’s worth pursuing a brand new pipeline to move your supply from the origination point in Pennsylvania to the high demand market in and around Massachusetts’s state capital, located about 500 miles to the northeast. But would the benefits of a new pipe outweigh the costs over the long term? We ran the model for a 1.0 Bcf/d, 500-mile pipeline to find out how big it needed to be. Assuming a typical operating pressure of 1,440 psi and flow velocity of 20 mph, we concluded that you’d need a 30-inch pipe (to see a detailed discussion of the calculations involved in coming up with that, refer back to Part 2).

And then in Part 3, we completed Steps 2 and 3 of the model—figuring out how much a 30-inch hypothetical pipe might cost to build and (much more important to you as our hypothetical producer-shipper, what the pipe might charge shippers to move gas) the daily transportation rate per MMBtu, based on data collected for recent projects in the Northeast. We measured the “inch-miles” of the pipe, the distance times the diameter, and then used a typical Northeast unit cost of $225,000 per inch-mile to give us an estimated capital cost of $3.4 billion to build the pipe. Then using a first-year effective annual cost factor (at typical levels based on decisions and settlements at the Federal Energy Regulatory Commission, or FERC), we calculated the likely tariff rate the pipeline developer might charge under a couple of different scenarios—depending on its tolerance for financial risk. We also estimated what the developer might offer in a negotiated rate, the most common way of capturing a new market. We found the rate could be as high as $1.81/MMBtu and as low as $1.15/MMBtu. So as long as the price differential to the destination is expected to remain higher than that transportation cost, we can conclude that the pipeline is worth pursuing.

Now that we’ve gone through the basics of the model, we want to do a couple of things: 1) see how it corresponds to the real world, and 2) to see how much volume needs to move on the pipe to get the most bang for the buck. To do the first, we ran an actual planned project through the model. To do the second, we just ran the model for different volumes and examined the trends of the results.