Author: Brynne Kelly (w/Lee Taylor technical levels)
The curve structure in oil has been devoid of direction for several months, with curves shifting in parallel up and down, providing little insight about the next big move.
Not much has changed in this regard during the first three weeks of September. The month began with the 12-month continuous WTI futures strip steadily above the $40 level, then promptly proceeding to drop below $40 in the front few months by mid-September. An Atlantic Ocean full of storm activity and a sizable draw in EIA inventories were enough to lift the curve back above $40 by the end of last week (yellow-green line below).
What's interesting though is the relative strength that WTI markets have shown within the petroleum complex this month as spreads across the board continued to trim any fat left on the bone. Any meat left on the LLS vs WTI spread was trimmed to the bare minimum (left graph below, LLS/WTI spread). The collapse in LLS/WTI is of course driven by the tightening of the WTI/Brent spread. WTI gained in value relative to its European counterpart (right graph below, WTI/Brent spread). The yellow-green line in both charts represents September 18 futures settlements for the continuous 12 month contracts.
WTI also gained value relative to NYH ULSD futures, which caused the HO/WTI crack spread to sell off (left graph below). Consequently, US ULSD lost more value during September than WTI crude oil did. Only US RB gasoline markets held their ground and actually gained value against WTI (right graph below). In fact, gasoline is the only market at the moment in which we are able to see tangible evidence of demand outstripping supply.
This may seem counter-intuitive to the 'work-from-home-no-one-is-driving' narrative that has been so prevalent in the market's psyche since the pandemic hit. But, the simple fact is that with US refiners running at significantly reduced capacity for the last several months, the US has been consuming more gasoline than it is producing. This is evident in EIA gasoline draws from their record-high levels back in April. This is manufactured 'tightness' driven by restraint in refinery output. This tightness does exist though, manufactured or not and it's the only real point of contention in this market that could lead to something.
To see this more clearly, we first needed to calculate a proxy for US gasoline demand, called 'Implied Demand'. To do this we used 3 common numbers reported weekly by the EIA:
Finished Gasoline Supplied by Refiners
Net US Imports of Gasoline
Change in US Gasoline Inventory
We assume that weekly demand for gasoline is filled first with 'production' (gas supplied by refiners) and net imports of gasoline. If demand exceeds production plus net imports, the difference is made up by a withdrawal from storage. Conversely, if production plus net imports exceeds demand, the difference is made up by an injection into storage. Timing differences within the report itself should smooth themselves out over time. Timing differences are why 'Implied Demand' (shown in the blue shaded area below) dipped below zero on two occasions in the last 5 years. Regardless, we see the trend: gasoline demand reaches it's lows in the US around December/January.
Gasoline supplied (produced) by refiners on a weekly basis is represented by the cyan line above. Given the static nature of refining capacity, it's no surprise that gasoline output is fairly constant over time. As a result, storage (and to a lesser extent, net imports) plays a very crucial role in meeting weekly and seasonal swings in demand. This dynamic is ultimately expressed via 'price' which is represented by spot prices (black line above).
In the next chart we isolate spot gasoline prices vs weekly EIA inventory levels. The picture is less clear in this 2-dimensional comparison - the relationship of price to inventory levels seems more random without including demand.
Overlaying weekly inventory levels on top of our initial chart we can see that gasoline prices react to anticipated changes in demand differently at different inventory levels.
This brings us back to our earlier point that gasoline markets have displayed the most relative strength this month. The market is reacting to the tightness created by refinery output falling short of implied demand. The clock would appear to be ticking on that sentiment if history is any guide. The market knows that demand is fast approaching seasonal lows in demand.
The next 3 months will be key. Will demand continue to outpace refining output at a rate that brings US inventories to seasonal low levels? A lot of progress towards this end has been made since April, but it must continue through year-end or gasoline cracks will come under significant pressure.
Curve Structure
While intra-commodity spreads have been under pressure in favor of WTI, of note this week is the slight crack in the middle and long-term WTI calendar spreads. Futures curves can be broken into 3 broad groups: Short, medium and long-term based on their related economic influences. In the short-term (less than 12 months), the curve is affected by supply, demand and storage availability. These factors are dissected daily and the curve responds to changes in these short-term fundamentals. The medium-term curve (between one and three years out) is affected by macro-economic fundamentals, which include hedging decisions and global economic forecasts. Finally, interest rates, financing-related structures/investment decisions and arbitrage activity affect the long-term market
Since the beginning of September, the short-term curve (12-month continuous strip) has shifted lower, but the curve structure is relatively unchanged as of last Friday's close. If anything, there is a slight flattening of contango between some of the middle months, but the 12-month spread is basically unchanged.
Over this same time period, the contango within the middle and long-term parts of the curve has steepened slightly (yellow-green line = 9/18/2020 CME futures settlement prices).
The tail end of the 2026 strip briefly settled above $50 earlier this month, but was quickly rejected. That leaves us with a sub-$50 crude strip for the next 6+ years.
This steepening of contango within the middle and long-term curves could suggest the entrance of hedging activity in the calendar 2021 & 2022 part of the curve. Bottom line, without some meaningful flattening of the curve structures, oil prices will struggle to find support and continue to shift higher and lower in parallel form. Increased hedge activity (selling) in the middle of the curve will either help to flatten the structure or spook the front of the market and lead to another round of selling in the front of the curve.
EIA Inventory Statistics
Weekly Changes
The EIA reported a total petroleum inventory DRAW of 6.50_million barrels for the week ending September 11, 2020. Commercial Crude oil inventories posted a weekly DRAW of 4.30.
Year-to-date, total Inventory additions stand at a BUILD of 95.90 million barrels (vs 99.30 last week). The YTD Commercial crude oil inventory build is now less than they were for this week-ending in 2015!
Commercial Inventory levels of Crude Oil (ex-SPR) and Refined Products are slightly above the 5-year average in oil, slightly within the 5-year average for gasoline and still way above the 5-year average for distillate.
Lee Taylor - Technical Levels
BRENT
Resistance: 43.76 / 45.08 / 46.50
Support: 42.06 / 39.32 / 36.98
The Brent market did hold the 39.63 - 39.26 area and is technically mirroring the WTI market. We still maintain that we need to see November Brent break above 42.91 to get extremely bullish. A break and settle under $39.26 in November Brent would make me reevaluate any length I have. A settlement above 43.76 in Nov Brent would signal a move back up to the August highs of 46.61. As I stated last week, the longer Nov/Dec Brent stays below -.59 the better chance of reaching our objective of -.71.
WTI
Resistance: 41.22 / 43.05 / 43.81
Support: 40.33 / 39.44 / 36.58
What a difference a week makes! We were saying last week that the market was oversold, and it would have a difficult time breaking under the $36.08 level basis for October WTI. This week would slide our attention to November with October going off the board on Tuesday. November WTI track was similar as it bounced off the 38.2% retracement (36.58) from the Covid-19 collapse. We would like to see Nov WTI hold 40.33 and regain some steam to the upside. Nov/Dec WTI did bottom but now seems to be right at the 50% pivot of -30. Strongly suggest keeping an eye on the -30 level as to how you play Nov/Dec.
RBOB
Resistance: 1.2545 / 1.2765 / 1.2941
Support: 1.1681 / 1.1316 / 1.0290
We got the settlement above 1.1681 in October RBOB for a retest of the 1.30's. October needs to hold the 1.1681 level to have any chance at a rally. The first real resistance area is 1.1341. We are still focused on Dec/March RBOB as it has formed a massive base near the -500 level. Despite its’ rally last week, still would look at adding length on a pullback to -371.
HEATING OIL
Resistance: 1.1648 / 1.1848 / 1.2090
Support: 1.1306 / 1.1037 / 1.0763
Heating oil rallied nicely late last week and now uses 1.1306 as major support. October heating oil's next objective is 1.1848, and if it can settle above then 1.3606, it can be a reality. We continue to be bullish heat spreads and if Nov/Dec heat can settle above -174, our objective becomes -101. We are still are liking Dec/March breaking above -500 to -445 then -396.
Comments