Some Clarity On Oil Storage Capacity Estimates
People are probably a bit confused to see constant references to full storage tanks, and then data from the Energy Information refers to U.S. crude storage capacity utilization being less than 60% while on a global level, the IEA puts it at 63%.
Available global oil storage capacity is variously put at 0.9, 1, 1.6 and 1.8 billion barrels, and the IEA’s estimate of capacity at 6.7 billion barrels would imply 2.5 billion barrels of available capacity. The contango in the market, which is extreme for the first three months, suggests that the cost of holding oil is very expensive, which certainly supports the ‘full tanks’ thesis, at least up to a point.
Much of the confusion represents uncertain data, but also operating conditions and terminology. First and foremost, capacity is not always capacity, and empty capacity is not always available. Storage might be under contract but currently empty, meaning it cannot be immediately utilized, as in the current market, most would be reluctant to give up leased capacity except at elevated prices. Additionally, it is not possible to mix ‘clean’ products like jet fuel, gasoline, and diesel fuel as the contamination makes them worthless. Different crudes can be mixed (and often are blended intentionally) but not always efficiently.
(It should be noted that the crude oil inventory data shown here are different from those published in the EIA’s summary of the weekly Petroleum Status Report, which includes pipeline fill and stocks in transit.)
Most important, the distinction between shell or nameplate capacity and working storage capacity can lead people astray. In the U.S., the difference is roughly 100 million barrels; this helps explain some of the apparent discrepancy between anecdotes of full tanks and reports of inventory being well below capacity. Shell or nameplate capacity usually reflects the maximum theoretical amount that equipment can handle, such as a refinery or pipeline. Think of it as the flow rate in the largest section of a pipeline, whereas working capacity would be the flow rate of the smallest diameter section. The usable capacity is thus the flow rate of the smallest pipe.
The good folk at the Energy Information Administration have provided an excellent description of the storage capacity data and why the total volume of tanks is not the same as the working capacity. First, the oil at the bottom of a tank (called ‘bottoms’) cannot be easily accessed because suction becomes inefficient as the bottom is approached. (See the figure.) Second, there is some space at the top (not very much) that cannot be utilized, in part because changing temperatures can mean the volume of the oil can shrink or expand.
The crucial figure is the contingency space, which is needed because of uncertainties in deliveries requires an amount of reserve capacity to maintain flexibility. The precise amount is not clear, but overall, most assume that working capacity is roughly 80% of shell capacity. This explains why the reported global storage capacity of 6.8 billion barrels, with a 60% utilization rate, does not square with reports of only 1 or 1.6 billion barrels of available capacity.
But it also implies that “working capacity” is not an absolute amount and operators could, in theory, add something like 5-10% to available capacity by using some of the contingency space: if you know you are going to stop accepting crude, you don’t need the flexibility which the contingency space provides. This could mean an additional 300-600 million barrels of usable storage capacity, which is not trivial. If economic recovery begins by the end of May, it is quite possible that storage capacity will be less strained that many expect. And lower shipping of crude from OPEC+ will also free up tankers to be used for floating storage. That issue will be addressed later.