Enjoin This!
Legendary Chicago prop shop* DRW has sued the CFTC, asking a Federal judge to enjoin the agency from filing an enforcement action against the firm. Based on the facts alleged in the DRW complaint, CFTC should pray that the judge grants the injunction, thereby saving the agency of total humiliation at trial.
The details are kind of geeky, but I’ll try to summarize in as close to plain English as I can. At issue is DRW’s trading-and quoting-of swap futures contracts traded on the Nasdaq Futures Exchange and cleared on the International Derivatives Clearinghouse. As a cleared contract, the swap futures was subject to daily mark-to-market. OTC swaps (at the time) were not collateralized or subject to mark-to-market. As basically anybody who has been around futures markets since the early-90s knows, this differential margining treatment should cause swap prices to differ from futures prices.
Hell, this has been known in the academic literature since 1981, with the publication of a paper by Cox, Ingersoll, and Ross. The basic idea is that marked-to-market futures have payoffs that are linear in the interest rate, but swaps have payoffs that are a convex function of the interest rate due to discounting: futures cash flows respond immediately to interest rate changes due to daily variation margining, but the effects of interest rate changes on the swap are not realized until the expiration date of the swap. This leads to a “convexity bias”: the futures are biased relative to the swap, due to the fact that with discounting, the present value of the swap cash flows are a convex function of interest rates.
Basically, there’s an advantage to being short the futures compared to being short the swap. If interest rates go up, the short futures position profits, and the short can invest the resulting variation margin inflow at the higher interest rate. If interest rates go down, the short futures position loses, but the short can borrow to cover the margin call at a low interest rate. The swap short can’t play this game because the OTC swap is not marked-to-market. This advantage of being short the future should lead to a difference between the futures yield and the swap yield.
DRW recognized this difference between the swap and the futures. Hence, it did not enter quotes into the futures market that were equal to swap yields. It entered quotes at a differential to the swap rate, to reflect the convexity adjustment. IDC used these bids to determine the settlement price, and hence daily variation margin payments. Thus, the settlement prices reflected the convexity adjustment. Not 100 percent, because DRW was trying to make money arbing the market. But the settlement prices were closer to fair value as a result of DRW’s quotes than they would have been otherwise.
CFTC apparently believes that the swap futures and the swaps are equivalent, and hence DRW should have been entering quotes equal to swap yields. By entering quotes that differed from swap rates, DRW was distorting the settlement price, in the CFTC’s mind anyways.
Put prosaically, in a way that Gary Gensler (the lover of apple analogies) can understand, CFTC is alleging that apples and oranges are the same, and that if you bid or offer apples at a price different than the market price for oranges, you are manipulating.
Seriously.
The reality, of course, is that apples and oranges are different, and that it would be stupid, and perhaps manipulative, to quote apples at the market price for oranges.
The CFTC is completely confused. Shocking, I know. Perhaps-perhaps-trading to affect the settlement price is a necessary condition for a manipulation allegation. It is not a sufficient condition. The crucial issue is whether the trading/quoting moves the price towards the appropriate competitive price, or away from it. Prevailing law requires a finding of an artificial price (for a completed manipulation) or the intent to cause an artificial price (for an attempted manipulation). DRW’s trading-quoting, actually-moved the settlement price closer to the proper value, relative to OTC swaps. That is, its activities made the settlement price less artificial. Bidding/offering at the OTC swap yield would have been artificial because such quotes would not have reflected the substantive economic differences between futures and swaps. Due to convexity and the lack of any adjustment for it in the Nasdaq Futures contract, there should have been differential between the two prices. Bidding/offering at zero differential would have caused settlement prices to diverge from fair value relative to swaps.
In other words, DRW contributed to convergence of the settlement price to fair value relative to swaps. Manipulative acts cause a divergence between the settlement price and fair value.
Many futures contracts have been mispriced for a considerable period of time post-introduction,. Eurodollars were mispriced relative to swaps until people figured out the convexity issue in the early-to-mid-90s: the academic literature diffused into practice much more slowly back in those days. But even simple cash-and-carry relationships have sometimes been violated in new futures. Most notably, S&P 500 futures were mispriced for a considerable period too in 82-83, and maybe a little bit afterwards (I’ll have to go dig up the papers that provide the evidence of this). Were people that bid/offered/traded to take advantage of/arbitrage this mispricing manipulating? That would be an odd theory, since this trading made prices less “artificial”, i.e., closer to fair value relative to the underlying.
In a sane world-or at least, in a world with a sane CFTC (an alternative universe, I know)-what DRW did would be called “arbitrage” and “contributing to price discovery and price efficiency.” Voluminous CFTC materials (and testimony by CFTC officials before Congress) identifies these as legitimate, and indeed salutary, futures market activities.
Another shocking aspect of this story is that the two biggest firms on the other side of the contract from DRW-Jefferies and, yes, MF Global-didn’t understand the effect of convexity. Jefferies claims it relied on IDC’s representations that swap futures and OTC swaps were equivalent, and hence thought it was making an arbitrage profit as a result of trading the futures and swaps at a differential. In fact, it was the source of arbitrage profits because it was trading at a differential that was too small, relative to what it should have been to reflect fully the convexity effect.
Moreover, IDC should thank its lucky stars that DRW did nudge the settlement price through its quotes. If it had marked the futures to market at the OTC swap yield instead of the price reflecting the DRW quotes, MF Global’s position would have been substantially under margined, subjecting the clearinghouse to substantial loss (or, at least, substantial delay to recovery) in the aftermath of MF’s default.
I shake my head in amazement at all this. Especially at the CFTC. It stands there dumb and mute when major manipulations occur, and then leaps on trading that is manipulative only in its deluded imagination (and complete ignorance of things practitioners have known for 20 plus years). Its manipulation enforcement is a race between Type I and Type II errors. It reminds me of the old joke about the lawyer who said “I lost the cases I should have won, but won the cases I should have lost, so on average, justice was done.”
Well, it would be a joke, but this isn’t funny. Especially since the CFTC’s authority now extends far beyond the futures markets, which are important enough as it is, to the far larger OTC markets. Huge sums are at risk to the ignorant judgments of the CFTC.
The CFTC has apparently adopted the William Clayton Rule: “The word ‘manipulation’ . . . in its use is so broad as to include any operation of the . . . market that does not suit the gentleman who is speaking at the moment.” This is not a good thing, when the “gentleman speaking at the moment” exercises the power of the Federal government. Hopefully the judge will do DRW, the markets-and the CFTC-a favor by slamming on the brakes on this inane enforcement action.
*Not to be confused with a chop shop! True story. My first year at Chicago a gang ran a chop shop out of the garage next to the student housing building where I lived. Quite an adventure when the cops raided the place at 6AM one morning. Shots fired. Reminded me of the old Second City bit where the cop fires several rounds and then shouts “Halt!”
So Prof as the lunatic far right-driven government shutdown approaches, where are all your pompous posts denouncing how weak the Tea Party clowns are making America look with their threats of forcing a shutdown? Where is your hot air about how they are dishonoring America? Or does this just not show up on your Fox News feed?
Comment by ohlord — September 18, 2013 @ 1:22 am
CFTC staff is unprepared to regulate the markets. They struggle to understand fairly straightforward pricing concepts especially when cash and futures markets diverge. We continue to argue that the CFTC needs less lawyers and more technology and trading talent.
Comment by David Downey — September 18, 2013 @ 8:42 am
@ohlord. Wow. Amazing how many unfounded assumptions based on stereotyping you can cram into 3 sentences. First, I pay zero attention to Fox News, on the tube, the web, or Twitter or anywhere. Second, I do happen to disagree with the shutdown strategy, and the defund Obamacare strategy, but it has nothing to do with projecting American weakness, dishonor, etc. Apparently you believe it is somehow analogous to Obama’s zig-zagging followed by a complete capitulation to Putin. Hardly.
@David. Don’t I know it. This reminds me of a story from 1992. After Clinton won, Philip McBride Johnson (a former CFTC chairman) said something to the effect “now we can get rid of the economists and put the lawyers back in charge.” And so it has been ever since.
@David-a major problem, of course, is that real trading talent won’t work for gov’t pay scale, and wouldn’t tolerate the stultifying bureaucracy. Moreover, re technology, CFTC has had a terrible track record since its inception on IT. There are numerous GAO reports ripping the agency for its IT failures.
One of the most distressing things about Frankendodd is that it expands the scope for failure. Unfortunately, failure scales up.
Interest rates were close to zero during that time frame. Convexity correction should have been deminimis Thirty year swap rates were trading through thirty year Treasury yields at that time. An investor could have purchased Treasury bonds on repo, earned a higher yield and had positive convexity.
More going on here than meets the eye.
Comment by Barry — September 18, 2013 @ 11:10 am
Re swap convexity correction: I think it needs to be calculated not on the immediate repo rate but than and the implied forwards, which were not 0.
Comment by sotos — September 18, 2013 @ 12:48 pm
@Barry & sotos. The convexity correction depends on rate volatilities and correlations over the tenor of the swap.
Professor,
First, have long enjoyed (not though not always agreed) with this blog.
Second, I would like to know what volatility was used to compute the convexity adjustment and the impact of change in market value.
Third, CIR (or even the founders of modern finance theory, Eugene Fama, Robert Merton, William Sharpe, and Harry Markowitz) never imagined a situation where any rate with credit risk (i. e, the thirty year non-cleared swap rate) trades below the a rate with no credit risk (the thirty-year Treasury rate). Investors were willing to pay a premium to receive a rate on an instrument with no convexity and credit risk over an instrument with positive convexity and no credit risk (teabagger arguments notwithstanding).
No one disputes the convexity differences between swaps and futures. Prices at which transactions were executed shows that the market did not assess much value to this convexity difference in the period in question.
Comment by Barry — September 18, 2013 @ 1:37 pm
@Barry. 1. Thanks. 2. DRW did apparently release a white paper detailing the magnitude of the convexity adjustment. I am trying to track it down. 3. Re the issue of prices at which transactions were executed. As I noted in the post, even simple no-arb restrictions (e.g., cash-and-carry-arbitrage) are sometimes violated in new futures contracts, and for a considerable period of time. Jefferies apparently completely ignored it because they relied on IDC’s representations that the futures and the OTC swap were perfect substitutes. (Risk Magazine wrote an article on this at the time.) If Jefferies had been right, and the contract had been mispriced in its favor, it would have made money instead of lost a lot of money. I usually rely on market prices, but in a new, and relatively illiquid contract, mispricings can and do occur.
DRW has, for years, been pushing this oddball futures contract that supposedly has proper convexity. Their white paper is a justification for this non-standard contact. Having been around the industry a few years, this controversy sounds very much like a “don’t throw me in the briar patch” strategy. DRW gets publicity for their “ERIS” exchange and manages to look like a free market maven fighting the evil federales.
You can find a copy of their paper here
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1783798
Again, I have to roll my eyes at discussions of convexity correction on thirty years swaps that trade through the Treasury. An investor can pay fixed on a thirty year swap, go long a thirty year Treasury, have no interest rate risk and earn positive carry. And the gains on the Treasury can be monetized immediately.
Comment by Barry — September 18, 2013 @ 7:21 pm
@Barry. Here’s the white paper.
I have not had a chance to look at the issue and have nothing specific to contribute but have a couple of general observations.
Normally, every theoretical “no-arbitrage” relationship in the market results from some “cash-and-carry” type argument.
One side of the argument relies on some borrowing strategy, while the other side – on short selling. Both arguments use idealized assumptions – no barriers to borrowing and no penalty for short selling. As a result they arrive to some “full-carry” relationships. On top of it they disregard the bid-ask spread and associated liquidity issues otherwise no equalities can be established (instead, inequalities will be).
My thinking is that while the full-carry argument may sound theoretically plausible in practice it almost never holds – in some cases it breaks down more, in other cases less.
Comment by MJ — September 18, 2013 @ 10:47 pm
Yo, SWP! Where was that plain English you were talking about??
Comment by Howard Roark — September 19, 2013 @ 9:38 pm
@Howard-everything’s relative 😛 And note well. I only promised to make it “as close to plain English” as I could!
[…] Streetwise Professor has an excellent analysis of the whole thing. […]
Pingback by The Feckless CFTC Strikes Again | Points and Figures — September 20, 2013 @ 5:39 am
Real talk—>NY banks were outfoxed by an independent firm, instead of competing, they called their regulator.
Comment by Jeff — September 20, 2013 @ 5:47 am