top of page
  • bkkelly6

Do Inventory Levels Equate to "Price"?

Author: Brynne Kelly

Despite the wild moves across the petroleum complex this year, markets have settled in to levels that look very similar to those seen in past years. It's almost as if the market associates a certain price level with certain inventory levels (at least during times when a clear future direction is lacking). This appears to be where we are at - clinging to relationships that have existed in recent history, specifically those from 2016 & 2017 (years which are more reflective of the new shale production and US export era).

Let's start with EIA crude oil inventory levels. We know that we are sitting at record highs at the moment. But, in all fairness, 2020 levels aren't TOO dissimilar to the extreme levels we saw in 2016 (brown dot) and 2017 (cyan dot), which at the time both stood out as outliers.

To refresh, 2016 was a year where inventories built to record highs by the beginning of summer (in preparation for new export capacity installations) and were generally unable to draw down meaningfully the rest of the year. At the beginning of 2017, oil inventories in the US once again reached new record highs, but with the help of OPEC cuts and expanding US export capacity, were able to draw down significantly in the 2nd half of the year. The long-awaited bottleneck that was backing up supply had finally begun to be removed. This is, of course, the million dollar question for the 2nd half of 2020: How will the bottleneck that created such oversupply be removed?

Interestingly, if we look back over the same historical time frame at WTI continuous futures prices (2010-present), we see an equally tight grouping of the same 3 years at similar price levels; the highest inventory level of the 3 years (2020) receiving the lowest price (green circle), followed by the 2nd highest level (2016, brown circle) and the 3rd highest level (2017, cyan circle). Dare we say the market thinks that inventory LEVELS have a corresponding PRICE level??

Of course we can look back and see how 2016 & 2017 played out. Intuitively, the year with the largest 2nd half draw-down in inventory (2017) saw the largest increase in price by the end of the year. In both cases though (2016 & 2017), futures prices, at best, meandered slowly and grudgingly higher in a relatively unmeaning way. In fact, given all of the shocks to the oil complex since the financial crisis, yearly price trends have been fairly benign (especially when compared to those seen so far in 2020!).

If outright price levels attempt to find an ally in inventory levels, perhaps spread markets are where things will look different. For that, we first look at the spread between WTI and Brent month-1 futures below, after all, this spread is how US oil prices become connected to world dynamics.

Again we see a pattern, notably that higher inventory levels in the US tend to result in a tighter spread between WTI and Brent (except for the spike lower during the May-20 futures expiration debacle). Due to the extreme size of OPEC+ cuts at the moment, the 2020 spread comes in slightly wider than our 2016 & 2017 comparative years because supply outside of the US has been more restricted. Yet as we progress towards the end of the year, we see the the larger draw-down of 2017 inventories resulting in a wider WTI/Brent futures spread (from almost flat to negative $5.00 by year end) solidifying the US as a supplier rather than a consumer that needed to pay a premium for imports. 2016, in contrast, where US inventories remained high and exports were still plagued with bottlenecks, there was more uncertainty that resulted in WTI trading over Brent several times. Fast forward to 2020, market participants are definitely in search of which world market will display more relative supply/demand balance vs others. A tighter balance in the US could move WTI over Brent once again, and vice versa. The complicating factor here is the amount of capacity overhang that exists everywhere. Meaning, which ever region shows the first initial relative strength will be the recipient of aforementioned excess capacity

And it's not just US crude oil producers that are grappling with record-high inventory levels. The same can be seen in Natural Gas markets where production has also pushed inventory levels close to new highs.

Moving on from energy producers to energy refiners, we compare inventories and spreads of refined products. First up is US Gasoline. The excess is much more visible here on a comparative basis between 2020 levels vs those in 2016 & 2017.

Granted, the outright size of inventory storage capacity is much greater for crude oil, however the overhang is much more visible in gasoline.

Surprisingly though, when we look at gasoline cracks, 2020 spreads to WTI are sitting at levels much like they were in 2016. The overall volatility of the front gasoline crack was much greater in 2nd half 2017 though, which had the lowest inventory level in July of the 3 years.

Using only history as a guide, the 2020 gasoline crack appears much more vulnerable to the downside, given current inventory levels.

Finally, we make the same comparisons for US Distillate inventories and prices. We have a new contender in the mix to point out, and that is 2010. However, we are going to stick with our 2016 & 2017 comparative years since 2010 was fundamentally a much different market. Similar to Gasoline inventories, the overall distillate storage capacity is much less than that of crude oil, yet the 2020 anomaly is still extreme.

This seems to be having a much greater impact on relative distillate cracks to WTI vs all other years (below). With inventory levels at such extreme highs, it's tough to play out a typical seasonal rally into winter (as seen in 2016 & 2017) unless something changes drastically.

With historical comparisons in mind we turn to the future - specifically the winter 2020-21 gasoline and distillate crack spread strips.

US distillate cracks (black line) have been weak for most of 2020 and have only recently stabilized. Gasoline cracks (blue line) for the upcoming winter season, which is typically the lowest seasonal demand for gasoline, moved swiftly lower with crude oil in early April, but have since recovered (although not as much as crude oil futures themselves, green line).

Given the overall inventory picture laid out above, it seems that at the moment, crude oil markets have a bit more clearance to the upside (if so desired) than their refined product counterparts. In that way, only a move lower in oil would be supportive of crack spreads, which seem to have run out of steam lately and taken a back seat to crude oil. The few levers that remain to be pulled are an increase in refinery runs (aka, demand for oil) which would only lead to a rise in crack spreads if there was an even greater increase in product demand.

Reminder: The next release date of the EIA Drilling Productivity Report is Monday July 13, 2020. As mentioned last week, it is within this report or next that we believe we could find some game-changing data regarding the reaction of oil producers to a rebound in oil prices.


Weekly Changes

The EIA reported a total petroleum inventory BUILD of 4.70_million barrels for the week ending July 2, 2020. Crude oil alone posted a weekly BUILD of 5.70 (excluding SPR). The only notable bright spot was the draw in gasoline of (4.80) which exceeded that predicted earlier by the API.

Year-to-date, total Inventory additions stand at a BUILD of 167.50 million barrels (vs 162.80 last week). There have been only 4 weekly draws of total crude oil inventories this year!

Commercial Inventory levels of Crude Oil (ex-SPR) and Refined Products are high everywhere as discussed earlier (except those at Cushing).


Lee Taylor - Technical Levels


Resistance: 44.01 / 46.34 / 48.40

Support: 39.69 / 36.49 / 34.16

Perhaps this week will bring a little change technically to the marketplace. Brent was like the WTI market in that our support and resistance numbers remain in place. A break under the 39.69 early in the week could signal a quick $3.00 move down to 36.49. Resistance is easy to find as one can just focus on the gaps above – we find the same gap in the weekly chart and the daily September, as we do the daily continuation which is 43.97-45.17. Sep/Oct Brent did test -.20 as we envisioned, however, we are now looking for the -.26 to -.33 as support, if not a move to -.69 is in store.


Resistance: 41.63 / 42.17 / 43.03

Support: 37.54 / 36.35 / 34.70

There is still one gap left in the crude market which appears in the August daily chart between 41.63 to 42.17. We have discussed the gap ad nauseam to date, but every week it appears to get heavier, stronger and/or more burdensome to the crude oil market. Our support and resistance levels remain, but look carefully to the support numbers as a move back toward 36.35 seems imminent. We continue to give our guidance for the front two crude spreads in which Aug/Sep and Sep/Oct WTI will remain stuck in a range between -25 up to flat. We do not see that changing unless a major move in flat price occurs.


Resistance: 1.3048 / 1.3253 / 1.3548

Support: 1.2254 / 1.1980 / 1.0916

The gasoline market was the lone commodity that tested and even broke through one resistance or support level last week. Aug RBOB traded as high as 1.2974 on the back of the Bayway news last week. After selling off early Friday, it rallied again on more Bayway news. It is still early here but please keep an eye on one important number – 1.2942 is the 50% retracement level on the monthly gasoline chart. If the market fails 1.2942 for the month, could signal a retracement of 20 cents.


Resistance: 1.2583 / 1.2771 / 1.2986

Support: 1.1850 / 1.1328 / 1.0611

August Heating Oil did not do much last week, so we tightened up our support and resistance levels in accordance. The market does seem a little overbought technically so a test of the support numbers may be the in the works. Resistance seen here in Aug/Sep heat at -100 with a break of that level projecting up to flat. Support major below seen at -160; please keep in mind that Aug/Sep remains overbought based on the RSI model.

261 views0 comments


bottom of page