Statement of Commissioner Dan M. Berkovitz Regarding the CFTC Staff Report on the Trading of Nymex WTI Crude Oil Futures Contracts On and Around April 20, 2020
November 24, 2020
The Report issued today (November 23, 2020) by the CFTC Staff, “Interim Staff Report: Trading in NYMEX WTI Crude Oil Futures Contract Leading up to, on, and around April 20, 2020” (Report) is incomplete and inadequate. The Report fails to determine the cause of the unprecedented plunge in the price of the WTI futures contract and divergence from physical markets on April 20, the penultimate day of trading in the May contract. Rather, it provides a general recitation of economic conditions in the weeks and days leading up to April 20, and offers only aggregated statistics regarding trading on that day. Unfortunately, this Report does not provide the public with an adequate explanation for the extraordinary price collapse on April 20.
As the regulatory body responsible for ensuring the integrity and fairness of derivatives markets, the CFTC should provide an accurate analysis of the events that caused the sudden and extreme price movement on that one trading day, in a manner consistent with the requirements of the Commodity Exchange Act. By leaving out important facts and analysis, the “interim, preliminary observations” in the Report do not provide the public with a meaningful understanding of the events of that day and their implications for our markets.
The inadequacy and incompleteness of the Report stems from the limited scope of factors identified in the Report and the absence of any analysis of the effect of those and other factors on the April 20 WTI price. The Report identifies certain “fundamental factors that impacted supply and demand for domestic crude oil,” and certain “technical factors,” such as overall levels of open interest, high-level measures of liquidity and trading, and the operation of circuit breakers. The Report states that these fundamental and technical factors “coincided with” the extreme price movements of April 20, and “may have influenced” those prices. But correlation—and even more so, coincidence—does not mean causation. The Report acknowledges as much: “this Interim Report [does not] identify the root cause(s) of any price movement of the WTI Contract leading up to, on, or around April 20, 2020.” In the absence of a root cause analysis, it is not possible to draw any conclusions from the presence of a few coincident facts identified in the Report.
The Report also suffers from these specific omissions and deficiencies:
Failure to analyze the lack of convergence. On April 20, the price of the WTI futures contract disconnected from the price of crude oil in the physical market and the price of other derivative contracts. Convergence with the physical market was re-established on April 21, the day of the final settlement of the contract. The extreme divergence on the penultimate day of settlement represents a disconnect on that day from the forces of supply and demand operating in the physical crude oil market. A significant number of commercial market participants have contracts that are priced in whole or in part in reference to the final settlement price on the penultimate day of settlement, and any divergence of that final settlement price from the fundamentals of supply and demand in the physical market can be harmful to those market participants. The Report does not even mention, much less analyze, the causes of the significant disconnect of the WTI contract from the physical market or other derivative instruments.
Insufficient analysis of availability of storage at Cushing, Oklahoma. The Report references “anecdotal reports” “suggesting” that crude oil storage capacity was in short supply at Cushing, Oklahoma—the delivery point of the WTI contract—prior to the April 20 expiration. The Report also cites data from the U.S. Department of Energy’s Energy Information Administration (EIA) regarding the levels of storage at Cushing. However, the Report does not undertake any analysis of the actual storage situation at Cushing, or the ability of market participants to make or take delivery under the contract, leading up to and during April 20. Such an analysis is necessary to determine whether storage scarcity or any type of squeeze—intentional or natural—resulting from the levels of storage at Cushing contributed to the price collapse. Significantly, delivery issues did not disrupt or cause unusual trading activity on the final settlement day, which indicates that tank capacity or other delivery issues may not have been a driving factor of the price activity the day before either. The anecdotes cited in the Report and the EIA data about the levels of storage at Cushing are therefore insufficient to draw any conclusions as to the degree to which storage concerns contributed to the price collapse on April 20.
Failure to analyze role of Trade at Settlement (TAS) contracts.  The report identifies the very large number of TAS contracts traded on April 20 and the “well above average” number of TAS contracts traded at the maximum price differential. But the Report provides no analysis of the price effect of this trading. The potential for TAS trading to artificially affect the settlement price of a contract is well known; the CFTC has brought two enforcement cases based on the use of TAS to manipulate the prices of futures contracts. The failure to analyze the price effect of the extraordinary levels of TAS trading on April 20 is a material omission.
Failure to analyze “flash crash” in last 20 minutes of trading. The Report accurately notes that most of the price collapse of the WTI contract occurred in “a 20-minute period between 2:08 p.m. and 2:28 p.m. ET, when the May contract prices moved from $0 to -$39.55 per barrel, before reaching the all-time low of -$40.32 at 2:29 p.m.” The Report, however, provides only general market and aggregated order book data and no detailed analysis or insight into why prices fell so far, so fast during this period. Moreover, the price of the WTI contract the next day rebounded and the final settlement price of the May contract on April 21 was $10.01. The general supply and demand factors identified in the Report that were present on April 20 were similarly present on April 21. The Report provides no explanation of how these same “fundamental” factors could contribute to both a settlement price of -$37.63 on April 20 and a settlement price of $10.01 on April 21. Accordingly, it is necessary to examine whether some other factor or factors not identified in the Report are responsible for the sharp drop in price in the last 20 minutes of trading on April 20, and the extreme difference in settlement prices between April 20 and 21.
It is crucial for the Commission to fully understand the collapse in WTI crude oil futures on April 20, 2020, and to share that understanding with the public as soon as possible. However, the issuance of an incomplete preliminary Report is a disservice to the public, market participants, and small and large businesses that depend on a reliable crude oil futures benchmark for contract pricing, risk mitigation, and price discovery. The Commission should continue to analyze the events of April 20 and provide a complete and accurate report to the public as soon as possible.
 Report at 5-6.
 Report at 6; see also Report at 5-6 (“In summary, a variety of factors coincided leading up to, on, and around April 20, when WTI Contract prices fell from $17.73 per barrel at the beginning of the trading session to settle at -$37.63 per barrel that day. An oversupplied global oil market faced an unprecedented reduction in demand due to COVID-19 slowdowns and shutdowns, and the uncertainty over supply, demand, and storage capacity coincided with price volatility in the WTI Contract observed at historic levels that day.”) (emphasis added); Report at 42 (“These fundamental factors coincided with a number of technical factors related to trading and liquidity which saw the May contract trade and settle at negative prices on April 20.”) (emphasis added).
 The Merriam-Webster Dictionary defines “coincidence” as “the occurrence of events that happen at the same time by accident but seem to have some connection.” See https://www.merriam-webster.com/dictionary/coincidence.
 Report at 4-5, n.4.
 This list is not exhaustive, but highlights a number of significant deficiencies.
 EIA reports that the negative pricing was “mainly confined to the [WTI futures] market.” EIA, Low liquidity and limited available storage pushed WTI crude oil futures below zero (Apr. 27, 2020), available at https://www.eia.gov/todayinenergy/detail.php?id=43495.
 The Report also fails to put the EIA data in full perspective. The level of crude oil in the tanks at Cushing in mid-April 2020 was approximately 60 million barrels, which is 16 million barrels less than the working storage capacity at Cushing of approximately 76 million barrels. Tank levels at Cushing had previously exceeded 60 million barrels for extended periods. In particular, from December 2015 through June 2017 the crude oil in storage at Cushing generally exceeded 60 million barrels, without any detrimental consequences for the WTI contract settlement process. See EIA Weekly Petroleum Status Report, Table 4 ( Nov. 18, 2020), available at https://www.eia.gov/petroleum/supply/weekly/.
 A TAS order allows a trader to execute, at any time during the trading session, a transaction at a spread to the settlement price. See CME Group, Trading at Settlement (TAS), available at https://www.cmegroup.com/trading/trading-at-settlement.html.
See In re Optiver US LLC, No. 08-Civ-6560, 2012 WL 1632613 (Apr. 19, 2012); In re Shak, CFTC No. 14-03, 2013 WL 11069360 (Nov. 25, 2013) (consent order); see also Craig Pirrong, Derived Pricing: Fragmentation, Efficiency, and Manipulation, Bauer College of Business, University of Houston, at 10 (Jan. 14, 2019), available at https://streetwiseprofessor.com/2020/04/ (“The analysis . . . demonstrates that TAS contracts create trading opportunities with asymmetric price impacts. This suggests that TAS may therefore also create opportunities for profitable trade-based manipulation, and this is indeed the case.”); Paul Peterson, Trading at Settlement for Agricultural Futures: Results from the First Month, farmdoc daily (July 29, 2015), available at https://farmdocdaily.illinois.edu/2015/07/trading-at-settlement-for-agricultural-futures.html (“Over the years TAS has been associated with several efforts to artificially influence the daily settlement price through ‘banging the close’ and other forms of manipulation [citations omitted].”).
 Report at 12.