The big energy headline today is that the US implied gasoline demand fell below 2020 levels (on a 4-week average basis).
For those of us that actually leave the house and drive, you can likely recall that traffic wasn’t nearly as bad as compared in 2020. Yet, the Energy Information Administration (“EIA”) data appears to suggest that gasoline demand is now below that for July.
This report brought a lot more questions than answers. For starters, US gasoline storage remains at the lowest level in the last 5 years for this time of the year.
Second, the EIA monthly data for gasoline demand has been trending higher than the weekly figures since January 2022.
As you can see in the finalized monthly data, there is a possibility to see a large discrepancy between the weekly and monthly data.
Third, there’s a growing disparity between the monthly US implied oil demand in relation to the weekly.
Finally, despite all the craziness and volatility in the weekly EIA US oil data, US total liquids with SPR continue to trend lower, albeit at a slower pace.
The same thing could be said of US crude with SPR.
What does all this mean?
While we want to be as empirical and unbiased as possible, it’s very difficult to take this gasoline demand figure at face value. The monthly data suggests that EIA has been underestimating US oil demand since April, and the latest May figure supports that assertion. Then there’s the fact that despite the horrible demand figures, product storage in aggregate continues to decrease, and even in light of the demand weakness from gasoline, distillate, and jet fuel, 3 is showing the lowest absolute inventories over the last 5 years for this time of the year.
Again, we are not trying to make up some bull narrative and argue that demand is not weak. In fact, we were one of the first ones to point out that US oil demand was surprising to the downside. But at some point, you just have to call a spade a spade. We think demand is weak due to increasing prices, yes, there’s no doubt about that. But is gasoline demand below 2020? No, and so we are likely somewhere between 2021 and 2020 demand figures, but skew towards the upside.
Why do we care?
We wrote at the beginning of the year that the OECD demand is the difference between really high oil prices and just high oil prices. OECD demand is the delta between us drawing 1.5+ million b/d or 0.5 million b/d. For oil bulls, how US oil demand performs is a key component to figuring out how the rest of the world performs. If US oil demand is indeed underperforming this much, then we should see global oil inventories build like crazy.
To put into context, US oil demand is ~3% below that of 2021 on a 4-week average basis. If you extrapolate that to global oil demand, that’s nearly ~3 million b/d gone. If this was true, then we should be seeing global liquids go up like crazy.
But according to the latest satellite data from Kpler, this doesn’t appear to be the case.
Again, I’m not trying to make up some fancy bull argument. I’m simply showing you the data I’m seeing.
Then there’s the issue of refining margins. If demand was so horrid, then we should see refining margins plummeting.
That’s clearly not happening.
All of this just leaves us with far more questions than answers. And quite frankly, the market needs answers because it is in the process of trying to figure out where the lower band is. In order for that to happen, we need reliably good data in relation to the oil price to figure out where demand starts to bottom. Without it, we are going to be swinging from the highs to the lows and back up without a clear band.
Now there is a way to validate all this, but it will take time. We can wait for the monthly data to come out, but in order to validate the July data, we would have to wait until the end of September. The market isn’t going to do that, so in the interim, it’s likely to swing excessively to the low side until demand responds, and then rebound from there.
Sadly, I think that’s where we are headed because, without reliable data, there’s no other way for the market to figure it out. So it’s just going to push to both extremes until it gets the answer it wants.
And while I hate saying this, we need more data to confirm the latest data. Our view is that this EIA data is likely off, but only time will tell. In the meantime, there are far more questions than answers.