Excess crude and key product inventories are normal when allowing for seasonality and demand.
Gasoline and diesel supplied (consumption) look suspiciously low and product exports appear unusually high, but otherwise, the picture is as it has been on the consumption side, with refined product consumption running about 5% below normal.
Incentive-to-store analysis shows that market sentiment has been absolutely crazy over the last month and a half. On Sept. 27, the market perceived a huge crude supply deficit. This has reversed into the current modest surplus, per our analysis of the futures curve. This really should not happen, but clearly, it has.
US oil production remains flat at 13.2 mbpd, as it has been for the last eight weeks. I cannot find a similar eight week stretch in the data, which suggests that the EIA is winging it, as the EIA’s new software upgrades remain unreliable in reporting production.
The Brent Spread (Brent – WTI) has widened to $5 / barrel, which is normally consistent with US production growth of around 800 kbpd / year. US production growth is likely under-estimated, possibly materially so.
Overall, the EIA’s software upgrade remains buggy. The missing barrel count (red circle, the difference between the light and dark blue lines, graph above), instead of declining as the EIA had no doubt hoped (green circle), has exploded, and seemingly in a cumulative fashion. This is not unusual for software modifications, particularly in such rich, data-intensive and near-real-time systems as the EIA uses. It’s really the 8th wonder of the world, but the EIA will require a few weeks to iron out the kinks. Take the data with a grain of salt in the interim.