For the past two days we’ve slogged through the math of converting natural gas pipelines to crude oil service. Whether or not you spent the time to follow the details, you might have wondered – What’s the point? Some pipelines are candidates for conversion, some are not – right? It just depends on the classic real estate rule --- location-location-location. Well there may be more to it than that. A lot more. What if I told you that a typical liquids pipeline can generate four times the revenue of a gas pipeline of equivalent diameter and length. The crude pipe can move four times the energy. And clearly, four times the energy and four times the revenue leads to four times the fun of running and operating a pipeline. Right now a lot of gas pipelines are not having much fun. On the other hand, it is party time at most crude oil pipelines. Bust out the red solo cups. Let’s figure out just how much fun they are having.
Note 1: This is Part 3 of the series on gas-to-crude pipeline conversions - A time for Gas, a time For Crude. In Part 1 we started by looking at the math of pipeline capacities to answer the question – how much crude oil can you put down a natural gas pipeline. Then in Part 2 we took a deep dive into the cost side of the equation. Today we get to where the action is – the revenue side. If you did not read Parts 1 & 2, this blog won’t make much sense. Get the remedial at the following links: Part 1; Part 2.
Note 2: Readers pointed out that I had a bust in Part 1, PROTC#4 – The calculation was backwards. The right magic number of pipeline conversions for a 8” pipe is 35,000/50,000 = .70, or for a 16” pipe is 78,750/112,500 = .70. And it works for any other diameter. The number has now been fixed in the Part 1 text. Sorry if you embarrassed yourself at any parties. It doesn’t change any of the other conclusions. (Warning - Don't do blogging with math after a glass of wine.)
We’ll take this in three steps. (1) What kind of tariff rates will regional price differentials support in the gas and crude oil markets? (2) Compute the value of these differentials based on the capacity of our example pipeline, and (3) Compare the value differential to the cost of conversion.
Price Differentials and Tariffs
Over the past couple of years, basis differentials (i.e., regional price spreads) in the natural gas market have evaporated. The combined impact of $$$ billions in new pipeline construction in the 2008-2010 timeframe, plus the crash in spot prices has reduced the differentials between most natural gas hubs to the variable cost of moving gas between the hubs. For example, yesterday’s differential between Henry Hub and Transco Zone 6 (New York) was $.23/MMbtu. The spread between the Rockies and Ohio (the route of Rockies Express) was $.38/MMbtu. From Carthage, TX to Zone 3 in Florida was $.33/MMbtu. On average, the differential between producing areas and market areas has been running about $.35/MMbtu. We’ll use that number in our example. Granted most of the fully loaded tariffs on natural gas pipelines (reservation fee plus commodity rate plus fuel) are well above these differentials. But those tariff rates represent the long term transportation commitments, not the spot value of the capacity. The difference between those fully loaded rates and the spot differentials in the market today are why many long haul natural gas pipelines are facing de-contracting risk. In other words, customers not renewing contracts when they come up for renewal. That’s the reason a lot of gas pipelines are not having much fun right now.
On the other hand, crude oil differentials are either wide or blowing out. As we discussed here in A Perfect Storm in the Bakken and Bakken’ and a Rollin’ at Clearbook and Guernsey, crude oil differentials are wide from the Bakken to Cushing ($4.00/bbl yesterday but up to $25 a few weeks back), and wider still between Cushing and the Gulf Coast ($13.50/bbl yesterday up to $30 a few months back). This is the reason why it is party time for crude oil pipelines. Transportation rates may be capped at the filed tariff level, but wide differentials are very good things for the pipelines. They support high throughputs (i.e., the pipes are full), the pipes are fully contracted, and the differentials support economics for capacity expansions and new pipes. Of course not all pipelines move crude between these major trading hubs. So we’ll be quite conservative and assume a $2.00 differential for a typical producing area to refinery spread for our example pipeline. Why pick this number? First, it is pretty typical for what you hear in the market for our example 16”, 500 mile pipe. Second, a 16” pipe would generate $100MM/year at a $2 rate. At a 20% annual revenue requirement, this level would support something like a $500MM capital expenditure for a crude pipeline. Based on our estimates in Part 2, that passes the smell test.