Wednesday, May 29, 2019

US Oil Inventories To Tighten -- SeekingAlpha Contributor -- May 29, 2019


Later, 2:23 p.m. CT: see comments as well as this stand-alone post regarding HGLs.

Original Post 

I suppose we will see the US crude oil inventory numbers a day later than usual due to the Monday holiday. If so, the API data will come out today; the EIA data tomorrow, Thursday, May 30, 2019.

From SeekingAlpha, earlier this week, predicting EIA numbers this week, this is from HFI Research:
  • The tale of two oil markets is starting to get more attention from the sell-side community with two excellent reports from Raymond James and RBC (in the report).
  • Everything points to lower US crude storage, as Brent-WTI Houston shot up indicating higher crude exports and 3-2-1 crack spread indicating higher refinery throughput.
  • There was no SPR this week, so our estimate would be -3.4 to -5.4 mbbls for this week's EIA crude storage report.
  • Global oil-on-water is also trending lower, while floating storage is at the lowest level excluding Iran (30 mbbls) and Venezuela (15 mbbls).
  • We are projecting US crude storage to drop below 380 mbbls by year-end.
[Side note: note the abbreviation for "million." This is quite non-standard. Generally, volumes in this context are abbreviated "mm" for "thousand-thousand" or million.]

So, again, the contributor suggests a draw of around 5 million bbls when EIA reports later this week. But look at this: the contributor suggests that by the end of the year, US crude oil inventories will drop to below 380 million bbls. If so, that would be amazing.

A new term for me: "oil-on-water" -- oil in transit.

A new term for me: the "3-2-1 crack spread": The 3:2:1 crack spread approximates the product yield at a typical U.S. refinery: for every three barrels of crude oil the refinery processes, it makes two barrels of gasoline and one barrel of distillate fuel.

An old concept: floating storage.

Note: as long as we're looking at new terms, here's an acronym I had not seen before. Again, on twitter today, and it appears most folks have not heard of it; I had not: HGL, hydrocarbon gas liquids: link here.
From a supply-side perspective, HGL constitutes liquids produced at both natural gas processing plants (natural gas plant liquids, or NGPL) and refineries (liquefied refinery gases, or LRG). From a demand-side perspective, HGL is marketed as natural gas liquids (NGL) and refinery olefins. From a chemistry perspective, HGL includes the alkanes (paraffins) – ethane, propane, normal butane, isobutane, and natural gasoline (equivalent to pentanes plus and nearly chemically identical to plant condensates) – and the alkenes (olefins) – ethylene, propylene, butylene, and isobutylene. However, HGL does not include liquefied natural gas (LNG) or aromatics.

As defined above, HGL captures the complete set of gas liquids marketed by midstream and downstream companies. Production of HGL by natural gas processing plants, fractionators, refineries, condensate splitters, and other facilities can be surveyed and analyzed with consumption data to build a clearer HGL picture. Figure 1 provides a simplified view of supply, demand, and chemistry of HGL.

This taxonomy was developed to clarify which elements to include when retrieving data from EIA reports, to highlight refinery olefins, and to define more specifically NGL, LRG, and condensate terms used loosely in many publically-available reports.
For example, from a tweet:
The database shows approximately 3.2 million b/d of HGL pipeline capacity is also set to be added in just 2019. FYI - "HGL" is the EIA term for things like Propane, Ethane, Butane, etc.
Disclaimer: this is not an investment site. Do not make any investment, financial, job, travel, or relationship decisions based on what you read here or what you think you may have read here.


  1. Thanks for the HGL discussion. Has always sort of bugged me to understand it. I think field production is all alkanes (no double bonds) from natgas plants.

    Refinery production of alkanes is very small in comparison to NGPL. However, you can have some. I was at a refinery that had net production of propane, on an island, which lacked natural gas. In any case, refinery alkanes are not included in "total liquids" production like the numbers of IEA or OPEC. (I think.) This is because, since they are from a refinery, were already part of the crude numbers. Don't want to double count.

    Simple olefins (small molecules with double bonds, e.g. propylene as opposed to propane) are only produced in refineries. They are side products of some of the cracking and the like. They are not in natural gas at all So are not part of NGPL. Like refinery alkanes, they are part of supply but not of total liquids (since they were already counted when the crude was counted).

    Interestingly, refineries actually consume more NGLs than they produce. In particular, isobutane is desirable because it can be alkylated into longer molecules within gasoline and good octane branched ones. Norfmal butane desirable for same reason, either combined into longer molecules or transformed into iso (better since it has a branch) and then that is used. Natural gasoline is an extremely light liquid, also called "plant condensate" (but different from lease condensate). It's about 70 API degree "oil" and is actually a liquid at room temp. It's either mixed into gasoline directly or just put through the distillation tower as part of a blended crude.

    This link is a useful reference. Look at the first and third blue column and first grey column especially.

    I attached a version with annual numbers versus monthly ones because there are some strange monthly numbers (like "negative supply") that are seasonal. I think these are just physical contraction of stocks in winter, but I'm not sure 100%. But see this suggestive link for butane:

    1. Thank you. I've moved this over to a stand-alone post to make it easier to access, as well as encourage comments up and updates: