Most crude oil traders and investors are familiar with crude oil's seasonal return profile. For those unfamiliar, WTI's seasonality closely mirrors the seasonal pattern of refinery runs, which in turn mirror gasoline consumption trends. The year typically starts off slow as drivers limit driving to the essentials, before building into summer driving season, coinciding with a peak in refinery output.
After Labor Day, refineries undertake seasonal autumn maintenance coinciding with a drop in gasoline consumption, resulting in average 1.2 million bpd of refining capacity coming offline. This pattern of refining activity translates in crude price returns that look something like this over the long-term:
Does this mean shorting oil now is the high odds trade?
EquityClock.com data states that September, October, November, and December have 45%, 35%, 45%, and 60% odds of being profitable respectively. These numbers are averages, however, and there have been many profitable fourth quarters over the past 10-years. Looking at what separates the profitable from the unprofitable can be a helpful in determining what this year will bring.
Source: Adam Mancini
As you can see, 5 of the past 10 years saw positive returns for the September to December period. When it comes to distinguishing between the positive years and negatives, the key factor is weekly inventory levels relative to their 5-year average, a term oil analyst Art Berman calls comparative inventory. I use weekly crude oil inventories excluding SPR. By comparing absolute inventories to their rolling 5-year average, it is possible to see how inventory levels are performing independent of seasonal variation.
What I found was very clear. Between the end of August and the end of December over the past 10 years, oil inventories falling relative to their rolling 5-year average was always associated with rising oil prices, and vice versa. In fact, the correlation was -0.9, which is a nearly perfect negative correlation. Note: A negative comparative inventory change means that oil is falling relative to the 5-year average, or the gap between current inventory levels and the 5-year average is closing.
Here is the relationship:
Source: Adam Mancini, EIA.
In other words, if one expects oil inventories to fall relative to their 5-year average for the remainder of the year, one can expect oil to rise with a very high degree of certain.
What will the rest of 2017 bring for inventories?
I expect inventories to continue drawing strongly against their 5-year average for the remainder of the year. I discussed in a previous article entitled "2017 U.S. Inventory Draws Will Be The Largest In 15 Years" how the first part of 2017 saw the smallest cumulative build in oil in 15 years, and the year will end with a net draw that should be the largest since 2003 (and possibly longer).
Most importantly though, inventories have been falling relative to their 5-year average all year. Currently, crude inventories are 463 million barrels, or 80 million above the 5-year average (compared to 518 million barrels and 146 million barrels above the 5-year average in early February).
The EIA's STEO sees U.S. crude inventories ending 2017 at 471 million barrels, which would actually equate to a large build in comparative inventories (from a likely 83 million barrels at the end of August to 118 million barrels in December).
The EIA's STEO, however, has proven to be a poor predictor of production. It sees August inventories ending at 476 million barrels compared to last weeks data showing 463 million barrels.
Inventories should continue to draw
My expectation is the year will end with inventories sub - 430 million barrels, which would draw down comparative inventories. This will be driven by a few key considerations:
1) The EIA STEO sees U.S. production ending the year up 1 million bpd. This is likely optimistic, as it would involve a similar production growth pace to the H1 2017. This will need to occur despite the fact drilling costs have been rising year-over-year since March, and the pace of price increases is growing as 2016 drilling contracts expire and drillers gain pricing power. At the same time, while there are ample DUC's (resulting in optimistic growth forecasts), service companies are limited in their completion speed by labor shortages, and the effect of new inexperienced workers.
2) The Brent-WTI differential will drive stronger crude exports. The differential has been widening as OECD ex-US crude stock draws outpace OECD stock draws. This will result in increased Asian exports and will minimize the effect of autumn refinery maintenance.
3) H2 2017 demand will be 2 million bpd greater than H1 according to the EIA's latest STEO. This is well above the 1.5 million bpd H2/H1 average growth since 2013. European refineries returning from outages and Chinese refineries coming back online after a historic maintenance season should support this.
I welcome all comments and feedback. If you enjoyed this piece and would like to be notified of further articles, don't forget to click Follow at the top of the page.
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.