Noisy Data From EIA, Nothing More To It
EIA's oil storage report was weird to say the least with some major week-to-week swings that made even industry veterans scratch their heads. There are three items in the report that were notably off:
Massive drop in implied oil demand in other oils.
US crude imports jumped to ~8.3 million b/d.
The adjustment factor flipped from +1.326 million b/d to -1.151 million b/d.
On the crude side, EIA reported a higher build than we had expected (+3.7 million bbls vs +1.5 million bbls). What's more frustrating is not that the build was higher than expected, but the fact that this is the 2nd week in a row that our modified adjustment remains elevated.
The above chart shows the modified adjustment data since the beginning of 2023. And what you will notice is that following the introduction of "transfers to crude oil supply" in early November, our modified adjustment figure has actually averaged closer to zero. But this started to change as we got into March and the data has not improved since.
For those of you not familiar with our modified adjustment calculation, we take physical movements in tankers for both imports and exports, and we net it out against EIA's reported figures. Over time, these figures match up as it's impossible for the figures to differ for a long time (physical vs customs).
If there is a major disconnect over a long period of time, then it implies that US oil production is either understated or overstated. In this case, we think transfers to crude oil supply is slightly understated, which is jacking up the adjustment factor.
Consider that starting in February 2024, transfers to crude oil supply in weekly EIA oil storage reports dropped from +700k b/d last year to just +463k b/d. This automatically bumps up implied adjustment by ~240k b/d.
Regardless of our analysis of the situation, our modified adjustment almost never stays elevated for a long period of time. As a result, readers should expect this to reverse leading to a surprise in crude draws.