Why the Oil Prices Went Negative on April 20? - Barış Sanlı


Hedging is not a piece of cake. It may be a very dangerous tool since the biggest determinant is a future no one knows. Mexico’s hacienda hedge is world-famous, but the rest is not well known. We are not discussing how much refineries or airlines have lost in these contracts with covid19. We have the most powerful data, computers, and algorithms, but the risk is still a wild beast.


Commodity Futures Trading Commission (CFTC) has published an interim report on the negative WTI crude oil prices on April 20, 2020. Practically the slide started from 14:08 until 14:30. The report has not satisfied anyone, but it is a good read. I have not been able to correctly decipher all the technical parts. The message I get was, “small investors should not think of oil futures as just another investment.”

The biggest question was, who let the prices go negative? The simple answer is lack of liquidity and storage. The report underlines the signals before April 20. They also cite NYMEX warnings on negative prices. But “animal spirits” of investors thought that they are buying WTI cheap. In China, some of these investors had to pay banks money to settle their accounts.

The negative prices are not unique or not happening the first time. The report references an event at Texas: “natural gas dropped below zero in March 2019 as increased supply outstripped pipeline capacity near the Waha hub in the Permian Basin of Western Texas.”. In electricity, it cites the European power market.

Sometimes when I talk about oil pricing, I refer to it as a social construct. We define how the price forms, and it is not occurring naturally. The biggest clue we have is “under the NYMEX rule, and the May contract (and all months other than the designated active month of June) would settle on April 20, based upon the VWAP of the accumulated calendar spread transactions occurring between 2:28 p.m. and 2:30 p.m. ET” from the report. That specific time period is the key to settlement for the May contract.

There are circuit brakers, and they were triggered consequently. But there is a technical bit about which contract is coined “active contract.” During that transactions, the June contract was the active contract. So circuit breakers didn’t halt the non-active contract -May- transactions. The other important parameter is the amount of open interest volume at that time. There were larger than usual OI volumes in the market, that means there were more contracts traded for May, but while approaching settlement there were lots of not a settled-or rolled contract.

The report underlines this fact with the sentence: “May Contract’s OI at the start of the April 20 trading session was 108,593 contracts, approximately 69.4% higher than the trailing 12-month average penultimate day OI of 64,101 contracts”.


Then comes the most technical part that involves “non-reportable.” These are not usual, well-established investors but small investors or their representatives. Nonreportable’ position has been discussed in detail with graphs. When liquidity dried, these investors -some do not know that this contract was physically delivered- think they are buying oil cheap. If they could have stored the oil for one or two months, it will be a very profitable business, depending on storage prices. But storage and the landlocked location of Cushing were not the best place to find storage or transfer during a market panic.

My main message is two folds. The first one is how powerful these contracts are and how dangerous they can be. The second one is “the technical details are essential for the hedging instruments.” It was the lowest oil price in years, but many small investors who were thinking “they are buying cheap to sell in the future” have lost money. Oil price looks simple, and we all seem to know what it means. We have to recheck our assumptions.