top of page
  • bkkelly6

Storm-Induced Backwardation?

(if not now, when?)

Author: Brynne Kelly (w/Lee Taylor technical levels)

With two major storms heading for the USGC, market participants are frantically trying to figure out how to react. Does it matter? Is it completely out of place to think about the normal supply/demand influences that hurricane season brings? Does the pandemic override run-of-the mill storm fundamentals?

The fact is, the list of unanswered questions is long. Prior to two major storms heading towards the USGC within days of each other, the market was still trying to get it's mind around the path of demand recovery in light of reduced capital spending. Anticipation had been building around a weak dollar/strong equity/inflation narrative such that oil prices have flat-lined within a $3 range just above the $40 level without much conviction.

So, just as forecast models began to move beyond chaos theory towards stability, hurricane season rears it's ugly head. In any other period in history, back-to-back hurricanes in the USGC would be digested by a lens which views that as an historic anomaly.

We at Cornerstone Futures are not meteorologists, so we leave the specifics regarding Marco and Laura's paths to the experts. Today, we simply look at the facts both past and present in an attempt to provide some context. Specifically, we thought it would be helpful to provide some situational context. For that we use current times vs those that existed at the time hurricane Harvey rocked the USGC.

One of the key data sources people in the US turn to when a large storm enters, or is forecast to enter, the USGC is the Federal Bureau of Safety and Environmental Enforcement (BSEE). They provide a daily report of the amount of production being shut-in ahead of a storm.

As of August 23, 2020, the BSEE estimates that approximately 57.6 percent of the current oil production in the Gulf of Mexico has been shut-in. BSEE estimates that approximately 44.6 percent of the natural gas production in the Gulf of Mexico has been shut-in. This is in anticipation of Hurricane Marco and Tropical Storm Laura.

The BSEE states that "the production percentages are calculated using information submitted by offshore operators in daily reports. Shut-in production information included in these reports is based on the amount of oil and gas the operator expected to produce that day. The shut-in production figures therefore are estimates, which BSEE compares to historical production reports to ensure the estimates follow a logical pattern."

Shutting in production due to a storm used to be read as bullish, until people realized that there is often times an equal and offsetting loss of end user demand. But, wait...we have already experienced an unprecedented loss of demand due to the virus. This is the key difference. Does the market think that we can lose even more end-user demand than we already have due to these storms, or is demand at such low levels that a storm is almost viewed as 'stimulus'? As a nation we are already operating at minimal demand. People are already working from home and not flying. Is it possible that support and recovery efforts in today's storm world actually result in moving the dial in a positive direction? Is it possible that evacuations, staging news crews, and moving resources to areas in need actually look like an increase in end-user demand?

OK, let's look at some data.

As of the most recent EIA inventory report, PADD 3 (commonly known as the USGC area) has operable refining capacity of 9,983 mbpd but is only now running at 82.1% utilization. This utilization rate dipped well below 70% in April as pandemic-induced demand losses of refined products took hold. Out of this 9,983 mbpd capacity, a little over 55% of that is located in Texas.

Perhaps a relevant comparison is the 2017 Hurricane Harvey that caused widespread flooding and damage to the south Texas and Louisiana coasts. While Marco and Laura are not anticipated to be at Harvey's strength, the possibility of a one-two punch could cause significant flooding.

Comparative Stats - Hurricane Harvey

It's impossible to avoid the fact that the impending storms conjure up the inevitable comparisons to previous storms such as Hurricane Harvey. In that spirit, we thought it would be helpful to review some of the 'price action' prior to and after that storm made landfall..

Reminder: Hurricane Harvey was a devastating Category 4 hurricane that made landfall on Texas and Louisiana in August 2017, causing catastrophic flooding and many deaths. It is tied with 2005 's Hurricane Katrina as the costliest tropical cyclone on record

Refinery Utilization

One of the defining characteristics of hurricane Harvey was the impact it's flooding had on the refining complex Prior to the storm, US refiners were running at greater than 95% utilization. Once the storm hit, this dropped immediately below 80%. This drop was almost equal in magnitude to the drop in utilization we saw this past March when pandemic destroyed demand.

Previously, hurricanes and other storms led to a huge drop in end-user demand because regular social activity (such as driving to work, going to a concert) were canceled which led to a sudden drop in refined product demand (for example, gasoline). Looking at the chart above and thinking about how anemic demand is already, one wonders if these storms will have the same effect, just from lower levels. That seems unlikely.

Crack Spreads

In 2017, as news began to spread of the immense destruction that flooding had on the Texas refining complex, there was an immediate and severe spike in front month crack spreads, specifically in gasoline cracks. Given the severe drop seen in refinery utilization, that seemed initially reasonable. Refiners were not operational, crude oil was not being processed, yet gasoline was still needed. Yet, this was a very short-lived spike once loss of supply was met with loss of demand (sound familiar??).


To put the above 2017 spike in crack spreads into perspective, we compare 2017 EIA inventory levels of gasoline (left) and distillate (right) below. In both products, inventory levels were much lower in 2017 than they are now. Something to think about.

Outright Futures Prices

We first looked at crack spreads above because of the impact that Harvey had on refining operations. We now see how short-lived that spike was. So, maybe spreads were constant but outright prices rallied? NO. Below we see a similar pattern. A quick spike in gasoline prices, but barely a blip on the radar screen for oil or distillate (other than seasonal).

Front-month Calendar Spreads

By now, it's probably obvious that front-month calendar spreads also only reacted for a brief moment, but we still include them for reference. Below are the Sep/Oct and Oct/Nov calendar spreads for RB, HO and WTI CL. At the end of August when Harvey hit, September crude had already expired which is why the Sep/Oct WTI spread (black line, bottom chart) seems to have not reacted while the Sep/Oct spreads in both RB and HO made new highs. Regardless, we see a familiar pattern: Front-month product spreads rallied at the thought of refinery closures while front month oil spreads took a dive.


At the beginning of this article we presented the production loss being reported by the BSEE. How does this compare to GOM (Gulf of Mexico) production shut-in's reported during Harvey?? Below is a recap of the BSEE daily oil and gas curtailed production due to hurricane Harvey. Does it surprise you that they are much lower for Harvey than today's curtailments for Marco and Laura? By a lot. Only 428,568 bpd at the peak for Harvey and already 1,065,614 bpd as of today's report combined for Marco and Laura.

Over the next few days, we will get much more clarity on all of this data, but we wanted to provide some context to the numbers. In a world where everything has been turned upside down, we thought this historical perspective might provide an anchor.


Weekly Changes

The EIA reported a total petroleum inventory DRAW of 7.50_million barrels for the week ending August 14, 2020. Crude oil alone posted a weekly DRAW of 1.60 (excluding SPR).

Year-to-date, total Inventory additions stand at a BUILD of 128.6 million barrels (vs 136.10 last week). Still no need to make comparisons because YTD builds far exceed anything on record.

Commercial Inventory levels of Crude Oil (ex-SPR) and Refined Products remain elevated compared to prior years, with distillate inventories being the most worrisome ahead of the winter heating season.


Lee Taylor - Technical Levels


Resistance: 46.36 / 48.82 / 51.16

Support: 42.89 / 41.57 / 40.12

The Brent market has mirrored the WTI market technically speaking. It also failed to hold 44.99 on Thursday and drifted lower Friday. I am now looking for the market to test 41.57. It is obvious that I should have stuck with my original call of staying bearish Oct/Nov Brent as it finally traded down to -.60. Assuming my conviction is correct regarding flat price, we could easily see Oct/Nov Brent trend towards -1.41.


Resistance: 43.68 / 44.40 / 46.41

Support: 41.73 / 39.34 / 36.08

It appears that we may go a full two months trapped in a $5 range. I have been doing this a long time, but I think I must go back to when I started in the business back in the early 90s to find more uninspiring times. I stated last week that “if October Crude Oil can hold 41.73 on Monday and continue to follow the trendline north, it will have no choice but to break through the major resistance”. Unfortunately, it failed on Thursday. Unless something changes with the weather in the next 48 hours, the market is on the brink of retesting 37.58.


Resistance: 1.2800 / 1.3021 / 1/3390

Support: 1.2237 / 1.1956 / 1.1789

The gasoline market is still strong, and one needs to be quite careful while this storm approaches and affects the Gulf. I am trying to focus on October RBOB as Sep is expiring next Monday. If we focus on some of the outer spreads – I believe there is an opportunity where they are overbought. For instance, Nov/Dec at the 175 level, Dec/June at -1937 and Dec/March at -350.


Resistance: 1.2516 / 1.2799 / 1.3023

Support: 1.1994 / 1.1817 / 1.1689

We may look back to last Tuesday’s activity as the last attempt to the upside. In my opinion, October Heating Oil will begin to trend lower and test 1.1848. October Heating Oil tried to hold the line at 1.2590 but came off hard on Friday. Oct/Nov Heat has resistance up at -196 and support at -228. I commented last week that “if the balance of the energy complex is in the midst of a small rally – it’s not smart to bear up on heat; however, if I am mistaken on crude then we will see October heat test 1.1734”. See what happens when you give into the bullish side of your brain!

262 views0 comments
bottom of page