DRW Decision Presents Significant Interpretation of CFTC Authority

December 10, 2018

By Michael L. Spafford & Daren F. Stanaway

On November 30, 2018—nearly two years after the conclusion of a bench trial—Judge Richard Sullivan of the Southern District of New York entered judgment in favor of defendants Donald R. Wilson and his company, DRW Investments, LLC, holding that the CFTC failed to meet its burden of proof with respect to its market manipulation, attempted market manipulation, aiding and abetting, and control person liability claims under the Commodity Exchange Act (CEA).[1] In short, the decision makes the following points clear:

Intent to create an artificial price, and not just to affect price, is required to show manipulation.
If a legitimate economic rationale supports a trading strategy, there is no basis for manipulation liability.
It is not illegal to be smarter than another counterparty, or to take advantage of a flaw in the market, so long as a legitimate economic rationale supports the trading strategy.
A sincere belief that one’s trading strategy and bids are more reflective of legitimate forces of supply and demand than existing market rates, coupled with a plausible explanation of the strategy (that is, the ability to demonstrate that the strategy reflected one’s view of the market), may afford a defense to a manipulation claim.

Factual Background

The DRW case involved the IDEX USD Three-Month Interest Rate Swap Futures contract, which had a floating rate based on the three-month LIBOR rate, traded on a futures exchange, and was cleared by the International Derivatives Clearinghouse (IDCH).[2] IDCH determined the settlement price of the contract via a multi-step methodology based in part on bids and offers market participants placed for the contract during 15-minute daily settlement windows.[3] The daily settlement price was significant in that it determined which party—the long or short position holder—owed the other a margin payment.[4]

Pursuant to a phenomenon known as the “convexity effect” or “convexity bias,” a long position holder receives margin at a “relatively better time to reinvest.”[5] This is because investing in a high interest rate environment generally yields a better return than investing in a low interest rate environment, so mathematically, the ability to reinvest margin payments is more valuable to the holder of a long position than a short position holder (a long party receives margin payments when interest rates are higher, whereas a short party receives them when interest rates are lower).[6] This effect does not impact uncleared OTC swaps, however (because they do not require daily exchange of variation margin), so the economic values of the two swaps are distinct; a cleared swap has some additional value to a long party (and a parallel lesser value to the short party).[7] Some clearinghouses endeavor to correct for the convexity effect by adjusting margin calculations pursuant to a formula, often called the “Price Alignment Interest” (PAI), to minimize price distortion and “bring the price of the cleared contract into closer alignment with the price of the uncleared contract.”[8]

In 2010, Mr. Wilson assembled a team of DRW employees to research potential trading opportunities in cleared interest rate contracts, which included modeling of the IDEX USD Three-Month contract’s fair value.[9] This led DRW to conclude that the fair value of the contract was significantly higher than that of comparable OTC swaps, and that the contract was “mispriced” in relation to the OTC swap rate (because it did not adjust for the convexity effect), creating an opportunity for arbitrage.[10] Based on this market view, DRW began to submit bids for the contract during the settlement window, which in turn affected the contract’s settlement prices, particularly given the illiquid nature of the market.[11] In total, DRW submitted more than 2,500 electronic bids between January and August 2011, none of which were consummated, though all were “binding bids” that any counterparty could have accepted had they desired to do so.[12] DRW did come close to consummating a trade in February 2011; the counterparty ultimately backed out of the deal but paid DRW a litigation release of about $850,000.[13]

That failed transaction triggered an IDCH investigation, during which DRW explained the convexity effect and its belief that its bids were much closer to the true value of the contract than were corresponding interest rates in OTC swap markets because the contract did not adjust for the convexity effect.[14] The IDCH never requested additional information, did not require DRW to change its bidding practices, and took no action against DRW.[15] DRW ultimately unwound its positions in the contract in August 2011, for about $20 million.[16] The exchange delisted the contract a few months later, in December 2011, finding it unlikely to generate significant business.[17]

Court Findings

In entering judgment in defendants’ favor, the court specifically highlighted “several gaps—concessions, almost” in the CFTC’s case, including that the CFTC presented: (1) no evidence that DRW ever made a bid it thought might be unprofitable; (2) no credible evidence that DRW ever made a bid it thought could not be accepted by a counterparty; (3) no credible evidence as to the fair value of the contract at issue when DRW made its bids (the CFTC alleged that defendants made bids on the contract above fair market value, but the CFTC’s expert “gave no testimony as to the fair market value of the contract”); (4) no credible evidence that DRW’s bidding practices ever “scared off” potential market participants; and (5) no evidence that DRW ever made a bid that violated any exchange rule (a point the CFTC conceded in its closing argument).[18]

In assessing the four elements of market manipulation, the court found that defendants had the ability to influence the contract settlement price, but that the CFTC failed to prove the remaining elements (specific intent to cause artificial price; existence of artificial price; and causation of artificial price).[19] Specifically, the CFTC offered no evidence that settlement prices were artificial. Indeed, the court rejected the CFTC expert’s argument that settlement prices were “artificial” because they were premised upon DRW’s bids, rather than consummated trades.[20] The court also credited defendants’ argument that the actual fair market price of the contract at issue was much higher than DRW’s bids (DRW believed the contract price was too low and submitted bids they believed more closely reflected the fair market value).[21] In addressing intent, the court distinguished between intent to create artificial prices and “mere intent to affect prices,” rejecting the CFTC’s argument that any price influenced by defendants’ bids was “illegitimate” and “artificial,” regardless of whether the bids reflected fair market value or were designed to attract counterparties for future transactions.[22]

With respect to the attempted manipulation claim, the court reiterated that “the mere intent to affect prices is not enough; rather, the CFTC must show that Defendants intended to cause artificial prices—i.e., prices that they understood to be unreflective of the forces of supply and demand.”[23] The CFTC itself acknowledged that if defendants made bids with an honest desire to transact at the posted prices, they could not be held liable for manipulation or attempted manipulation.[24] The court concluded that the evidence proved “beyond the shadow of a doubt” that defendants “sincerely believed” the fair market value of the contract was higher than the bids they submitted, and thus believed their bids were more reflective of legitimate forces of supply and demand than existing market rates.[25] Accordingly, their bidding practices were designed to bring the settlement price closer to what they believed was fair market value, to attract more swap counterparties.[26] The court credited defendants’ explanations of their trading strategy and held that because a “legitimate economic rationale” supported defendants’ trading pattern, their trading could not afford a basis for liability under the CEA.[27] The court further held the fact that none of the defendants’ bids resulted in a consummated trade insufficient to demonstrate intent to manipulate, particularly where nothing prevented potential counterparties from hitting their bids and “reaping the rewards of the allegedly inflated prices” and one market participant in fact did almost consummate a trade with DRW.[28] The court also noted that there is nothing wrong with “recognizing the flaw in the contract” and “taking advantage of it,” because “[t]hat’s what markets are for,” and the “so-called price distortion decried by the CFTC was simply a more accurate assessment of the fair market value” of the contract.[29]

The court concluded with a powerful statement: “It is not illegal to be smarter than your counterparties in a swap transaction, nor is it improper to understand a financial product better than the people who invented that product.”[30] The court went on to state that defendants “didn’t need to manipulate the market to capitalize on that superior knowledge,” and no evidence suggested that they did so.[31] Further, after the alleged manipulation, other market participants and the CFTC’s Clearing Division came to acknowledge that defendants’ view of the market—that the contract was not the equivalent of an OTC swap—was correct.[32] Only the CFTC’s Enforcement Division “persisted in its cry of market manipulation, based on little more than an ‘earth is flat’-style conviction that such manipulation must have happened because the market remained illiquid,” which “is not enough to prove” manipulation or attempted manipulation.[33]

[1] CFTC v. Wilson, No. 13 Civ. 7884, 2018 WL 6322024 (Nov. 30, 2018). The Senate recently confirmed Judge Sullivan’s nomination to the Second Circuit. Tim Ryan, Senate Confirms 15 Judges for the Federal Bench, COURTHOUSE NEWS SERV. (Oct. 12, 2018), https://www.courthousenews.com/senate-corfirms-15-judges-for-the-federal-bench/.

[2] Wilson, 2018 WL 6322024, at *5.

[3] Id.

[4] Id.

[5] Id. at *4.

[6] Id.

[7] Id.

[8] Id.

[9] Id. at *6.

[10] Id.

[11] Id. at *7.

[12] Id.

[13] Id. at *7–8.