As I read the commotion yesterday on Kinder Morgan (KMI)(KMP)(KMR), something struck me. The commotion revolved around Hedgeye citing Kinder Morgan as their new short idea. Obviously, Kinder Morgan and other pipeline MLPs are so popular and thought so safe with their large yields, that this short idea generated a lot of talk.
What struck me was what I believe is the underlying insight that's going to drive Hedgeye's logic for the short, which will be explained in a report that should be out on September 10. What follows is the explanation of this insight.
The insight that follows isn't mine. I got it from reading Kevin Keiser's Twitter feed and following his line of thought. I believe that this insight is quite brilliant, that's the reason why I am writing about it.
As everyone knows, the increase in U.S. natural gas and oil production is being driven by the shale revolution, which consists of using horizontal drilling and fracking to unlock shale reservoirs that were previously uneconomic to explore. These new wells, however, have a defining characteristic: their production can start off being high, but it declines very rapidly over time, with 80% declines in the first year not being unheard off.
There was always a debate regarding these production declines and whether it was economic to drill those wells. With the decline in natural gas prices, E&P companies focused on wet plays, going for NGLs and crude which fetched higher prices, hence the rapidly rising U.S. crude production we've been seeing. The debate regarding the viability of these wells died off somewhat because when selling higher-value products, the wells can pay themselves off in a very short time. But still, when looking at an E&P company, one must be aware that to keep the same level of production, the company needs to keep on opening new wells at a considerable rate, since the old wells decline so rapidly.
This means that the maintenance capex for an E&P company reliant on shale for its production is bound to be high.
On the other hand, when buying pipeline MLPs, people think that they're buying something akin to toll roads. The oil goes through, the company gets a fee. This and the high yields is perhaps the reason why pipeline MLPs routinely trade at EV/EBITDAs in the 15s-20s. While these companies are expanding the capex is naturally high, but investors think that once capacity is in the ground, the capex would become incredibly small - just the cost of keeping those pipes running in good condition.
But not all pipelines are made the same, and that's where I believe Hedgeye's insight lies. You see, the very first pipes right after the wellhead, the "gathering system", should behave more like the wells themselves, than like the high-pressure, long-distance, pipelines one usually will think about. That is, the gathering systems will not behave like toll roads. (Illustration: Atlas Pipeline Partners 10-K)
Why not? Because when a company connects a well, it will get fees for the product that goes by, but remember, that product will decline mighty fast in the case of shale wells. Then much like the E&P has to open another well to keep production level, the gatherer has to connect another well to maintain throughput.
What this means is that the maintenance capex on the gathering system is not just the cost of keeping those existing pipes flowing. No, to keep throughput level the gathering system has to continuously keep on connecting new wells! And that is the insight: the capex to keep throughput level is higher, much higher, than just the capex to keep the pipes running.
And it gets worse. While a well will pay itself off in a very short period of time, if we account for most of the value of the product coming out, the pipe connecting it to the gathering system will collect just a fraction of that product's value. The pipe should be reliant on a much longer term flow of product to pay itself off. But as we saw, the flow won't be there because it drops so fast.
Therein lies the great doubt. The capex on gathering systems is monstrous and there's no talk of 80% depreciation rates in the first year to have been used to match volume with the production decline profile. On the other hand, there are parts of the gathering system (compressors, processing facilities, etc) that are shared by the wells, and the cost that needs to be recouped is just the connection to each individual wellhead. In any instance, this dynamic raises the doubt of whether pipeline MLPs relying greatly on gathering systems will be viable or just giant houses made of cards.
It's not always easy to know just how much an MLP relies on gathering systems for its revenues. However, the production profile of shale wells raises the risk that the gathering systems servicing them might be uneconomic and this reality might be overshadowed both by accounting (not enough depreciation leading to accounting profits) and by growth leading both to growing revenues and profits, but negative cash flow coming from ongoing massive capex (some of which will be to finance growth, but a lot of which will be going just to keep throughput level).
At the very least the unknowns created by this dynamic make it hard to invest in companies where shale gathering systems might represent a large part of the business. This is probably the insight that drives Hedgeye's negativity towards Kinder Morgan, but it can also affect other companies such as MarkWest Energy Partners (MWE) or Atlas Pipeline Partners (APL).