Oil Refiners Are A Value Trap

Includes: HFC, TSO, VLO, WNR
by: Paulo Santos

When trading refiner shares, one could be mistaken that following the price of gasoline, heating oil, or crude, by itself, might give some kind of advantage, in a manner somewhat similar to what I described in this article.

That would be a wrong assumption. Yes, refiner shares do follow something, but it’s not the price of the product alone. Instead, they follow the crack spread, which is basically the differential between what the refiner pays for crude, and the value of the distillates it obtains from refining it.

Although the precise mix of distillates each refinery gets from one barrel of crude varies with the quality of the feedstock, formulations, time of the year, refining technology and process and other variables, we can use a proxy ratio to get a gross approximation, usually in the form of an X:Y:Z relationship – with X meaning the number of crude barrels, Y the number of resulting gasoline barrels and Z the resulting heating oil barrels. The most common ratio is 3:2:1, so 3 barrels of crude yield 2 barrels of gasoline and 1 barrel of heating oil.

Knowing this, we can thus calculate the historical, and present, theoretical crack spread and its evolution. This gives us something like this (Using Brent, Gasoline and Heating Oil, front month future prices):

Expressed as a percentage of a barrel of crude:

Now, comparing the evolution of the crack spread to the shares of refiners like Valero (NYSE:VLO), Tesoro (NYSE:TSO), Western Refining (WSN) and HollyFrontier Corporation (NYSE:HFC), we get this:

This is still a rather noisy chart, even though some relationships are already evident. We can smooth the crack spread and stock quotes using a 5 week moving average, to get this:

What we see here, is that the crack spread expressed as a % of a barrel of crude leads the movement in the share prices of the refiners. We also see that lately, the refiners seem to have gotten ahead of themselves, especially given the recent plunge in the crack spread. There’s a reason for this, albeit temporary – it stems from the abnormal spread that emerged between Brent Crude and WTIC crude during 2011 – usually Brent is slightly cheaper than WTIC crude, but lately WTIC became as much as $18 cheaper than Brent. I calculated the above crack spread using Brent, because Brent and other crudes have become the worldwide benchmark, whereas WTIC crude has seen depressed prices because of a technical limitation In the physical settlement of WTIC crude futures. The relationship between the two crudes can be seen in this chart:

The WTIC/Brent spread

This recent phenomena is benefitting mostly the US refiners. Usually, WTIC would be a bit more expensive and the true worldwide benchmark, however for the past year WTIC has been trading significantly cheaper because of a technical issue that led to bloated crude stocks in the Cushing, Oklahoma delivery point where WTIC futures are physically settled.

Basically there was too much pipeline capacity built to take crude into Cushing, and not enough capacity to take it out. Presently, some pipelines are being reversed to alleviate the problem, but in the meantime, WTIC crude is trading cheaper than Brent, whereas gasoline, diesel and heating oil still mostly enjoy prices being set worldwide by production based on the more expensive Brent (and other crudes), which are the true crude benchmark right now. This makes for a temporarily bloated crack spread when measured in terms of WTIC crude, which correspondingly temporarily higher margins for the refiners I named.


The temporary anomaly in the WTIC/Brent spread favors the US refiners that can get feedstock at WTIC prices, but sell distillates at worldwide prices. However, we can see that recently the anomaly has started to fade, so the WTIC crack spreads will also be converging to those calculated with Brent. This means a collapse in spreads and refining margins, which is already ongoing.

Since the present Brent crack spreads are the lowest we’ve seen in the last 5 years, the earnings that these refiners will be putting out once exposed to them will also probably be much, much lower than presently expected and reflected in the estimates (at higher crack spreads, the refiners were showing earnings that were 1/2 to 1/3 of present forward estimates).

So, the forward estimates on the refiners mentioned are overstated and will most likely suffer strong downward revisions in the coming months. This could mean that the presently attractive valuations presented below, are most probably a value trap.

Finally, for those interested in a deeper look into crack spreads, I recommend the NYSE handbook on crack spreads.

Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.