JPM Admits CIO Group Consistently Mismarked Hundreds Of Billions In CDS In Effort To Artificially Boost Profits
Back on May 30 we wrote "The Second Act Of The JPM CIO Fiasco Has Arrived - Mismarking Hundreds Of Billions In Credit Default Swaps" in which we made it abundantly clear that due to the Over The Counter nature of CDS one can easily make up whatever marks one wants in order to boost the P&L impact of a given position, this is precisely what JPM was doing in order to boost its P&L? As of moments ago this too has been proven to be the case. From a just filed very shocking 8K which takes the "Whale" saga to a whole new level. To wit: 'the recently discovered information raises questions about the integrity of the trader marks, and suggests that certain individuals may have been seeking to avoid showing the full amount of the losses being incurred in the portfolio during the first quarter. As a result, the Firm is no longer confident that the trader marks used to prepare the Firm's reported first quarter results (although within the established thresholds) reflect good faith estimates of fair value at quarter end."
As a result of this, regulators who now are only 3 years behind the curve, are most likely snooping to inquire not only how JPM did it (call us: we can brief you in 2 minutes), but who else has been doing this? Hint: everyone.
Because in other words, we have just discovered that the two key components of the entire CDS market: the LIBOR base and market "marks" have been bogus at best, and realistically, fraud. And one wonders why no bank ever will let CDS trade on an exchange...
On July 13, 2012, JPMorgan Chase & Co. reported that it will restate its previously-filed interim financial statements for the first quarter of 2012. The restatement will have the effect of reducing the Firm's reported net income for the 2012 first quarter by $459 million. The restatement relates to valuations of certain positions in the synthetic credit portfolio of the Firm's Chief Investment Office. The Firm's year-to-date principal transactions revenue, total net revenue and net income and the year-to-date principal transaction revenue, total net revenue and net income of the Firm's Chief Investment Office ("CIO") will remain unchanged as a result of the restatement. The Firm reached the determination to restate on July 12, 2012, following management review of the matter with the Audit Committee of the Firm's Board of Directors on the same day.
The restatement results from information that has recently come to the Firm's attention in connection with management's internal review of activities related to CIO's synthetic credit portfolio. Under Firm policy, the positions in the portfolio are to be marked at fair value, based on the traders' reasonable judgment as to the prices at which transactions could occur. As an independent check on those marks, the CIO's valuation control group ("VCG"), a finance function within CIO, verifies that the traders' marks are within pre-established price testing thresholds around external "mid-market" benchmarks and, if not, adjusts trader marks outside the relevant threshold. The thresholds consider market bid/offer spreads and are intended to establish a range of reasonable fair value estimates for each relevant position. At March 31, 2012, the trader marks, subject to the VCG verification process, formed the basis for preparing the Firm's reported first quarter results.
However, the recently discovered information raises questions about the integrity of the trader marks, and suggests that certain individuals may have been seeking to avoid showing the full amount of the losses being incurred in the portfolio during the first quarter. As a result, the Firm is no longer confident that the trader marks used to prepare the Firm's reported first quarter results (although within the established thresholds) reflect good faith estimates of fair value at quarter end.
The Firm has consequently concluded that the Firm's previously-filed interim financial statements for the first quarter of 2012 should no longer be relied upon, and the Firm will be filing an amendment to its Quarterly Report on Form 10-Q for the quarter ended March 31, 2012, as soon as practicable, but not later than it files its Quarterly Report on Form 10-Q for the quarter ended June 30, 2012. The financial statements included in the amended Quarterly Report on Form 10-Q will reflect adjusted valuations of the positions in the synthetic credit portfolio as of March 31, 2012, based upon external "mid-market" benchmarks, adjusted for liquidity considerations. While there are a range of acceptable values for such positions, the Firm believes this approach represents an objective valuation and is reasonable under the circumstances.
As a result of the restatement, the impact of the trading losses related to the synthetic credit portfolio on the Corporate/Private Equity sector during the first quarter will increase, as noted in the table above, but this increase will serve to reduce the impact of these losses on the Corporate/Private Equity sector during the second quarter by a corresponding amount. Accordingly, as noted above, CIO's year-to-date principal transactions revenue, total net revenue and net income and the Firm's year-to-date principal transactions revenue, total net revenue and net income will remain unchanged by the restatement.
The valuation errors had an immaterial effect on the Firm's balance sheet. CIO's Value at Risk model used, as inputs, independent marks for a majority of the positions in the synthetic credit portfolio and daily trader marks related to a limited number of positions in the portfolio. The Firm believes that if CIO's VaR were re-calculated for the first quarter of 2012, the re-computed CIO VaR numbers would not be materially different from those reported in the Firm's Quarterly Report on Form 10-Q for the 2012 first quarter. At June 30, 2012, average VaR for CIO was $177 million for the quarter then-ended, and was $153 million for the six months then-ended. For the Firm, average total VaR was $201 million for the quarter ended June 30, 2012, and was $186 million for the six months ended June 30, 2012. For the same reason, the Firm believes the valuation irregularities had an immaterial impact on the Firm's risk-weighted assets. However, as a result of the restatement, the Firm's Basel I Tier I common ratio will be reduced by 4 basis points to 10.3% and its Estimated Basel III Tier I common ratio will be reduced by 3 basis points to 8.1%, at March 31, 2012.
Management has determined that a material weakness existed in the Firm's internal control over financial reporting at March 31, 2012. During the first quarter of 2012, the size and characteristics of the synthetic credit portfolio changed significantly. These changes had a negative impact on the effectiveness of CIO's internal controls over valuation of the synthetic credit portfolio. Management has taken steps to remediate the internal control deficiencies, including enhancing management oversight over valuation matters. The control deficiencies were substantially remediated by June 30, 2012.
Management's internal review of these matters is ongoing. If the Firm obtains additional information material to its periodic financial reports, it will make appropriate disclosure.
Next up: we learn that just like Lieborgate, so was everyone else doing just what JPM admitted to doing as well.
In other news, there goes the entire CDS market.
* * *
For those who wish to learn more on this topic, here is what we said two months ago, predicting to the dot, all that was just confirmed above:
The Second Act Of The JPM CIO Fiasco Has Arrived - Mismarking Hundreds Of Billions In Credit Default Swaps
As anyone who has ever traded CDS (or any other OTC, non-exchange traded product) knows, when you have a short risk position, unless compliance tells you to and they rarely do as they have no idea what CDS is most of the time, you always mark the EOD price at the offer, and vice versa, on long risk positions, you always use the bid. That way the P&L always looks better. And for portfolios in which the DV01 is in the hundreds of thousands of dollars (or much, much more if your name was Bruno Iksil), marking at either side of an illiquid market can result in tens if not hundreds of millions of unrealistic profits booked in advance, simply to make one's book look better, mostly for year end bonus purposes. Apparently JPM's soon to be fired Bruno Iksil was no stranger to this: as Bloomberg reports, JPM's CIO unit "was valuing some of its trades at prices that differed from those of its investment bank, according to people familiar with the matter. The discrepancy between prices used by the chief investment office and JPMorgan’s credit-swaps dealer, the biggest in the U.S., may have obscured by hundreds of millions of dollars the magnitude of the loss before it was disclosed May 10, said one of the people, who asked not to be identified because they aren’t authorized to discuss the matter. "I’ve never run into anything like that,” said Sanford C. Bernstein & Co.’s Brad Hintz in New York. “That’s why you have a centralized accounting group that’s comparing marks” between different parts of the bank “to make sure you don’t have any outliers,” said the former chief financial officer of Lehman Brothers Holdings Inc."
At this point, Zero Hedge assumes that Iksil was merely abusing the little loophole used by every CDS trader since time immemorial, which however on a TRSed position of $100 billion in notional, which based on our calculations has a DV01 of $200 million, means that the bid/ask spread itself is worth $500 million in profit (and not so much loss).
However, if what Bloomberg is implying is that Iksil was effectively overriding "real" marks, and using imaginary (or "forced") bids and asks, then that brings into question the validity of CDS marks reported by MarkIt, the same MarkIt partially owned by Goldman and... that's right, JP Morgan (more on MarkIt in a moment).
But first, back to Bloomberg:
Jennifer Zuccarelli, a spokeswoman for New York-based JPMorgan, declined to comment on whether the CIO and investment bank were using different prices.
“All components of the synthetic credit portfolio in the chief investment office were mark-to-market,” she said.
The trades in question, made by a CIO group that included Bruno Iksil, nicknamed the London Whale because his positions grew so large, were on so-called tranches of credit-swap indexes, the people said.
Tranches allow investors to wager on varying degrees of risk among a pool of companies. Credit swaps pay the buyer face value if a borrower fails to meet its obligations, less the value of the defaulted debt.
Because JPMorgan had amassed such large positions, even a small change in how the prices were marked may have generated a big difference in the value of the trades, one of the people said.
While very little is known at this point, the realization that JPM did in fact abuse mark-to-market of a Level 1 security means that if Iksil's book was marked fairly, to mid-market alone, let alone to the real exit level opposite where it is most profitable (i.e., long risk as in the case of Bruno Iksil's IG9 holdings -> mark at offer, and vice versa), the losses would be materially greater, potentially up to hundreds of millions in the remarking process itself? And any further uncertainty about JPM's accrued losses, which we now know had to be covered up by tens of billions in asset sales from its portfolio (but, but JPM certainly did not need the cash) will merely add to the toxic spiral that is the pounding of JPM stock, coupled with further widening in IG9-10, which leads to even more JPM stock losses, which further blows out IG9-10 and so on.
One thing we do know is that in a recent case of a UBS prop trader, caught mismarking his CDS book, there was some serious litigation involved, and major accusation of illegalilty. Once again, from Bloomberg:
Ramon Braga, a trader on the bank’s corporate-credit desk in London, was fired for collusion in the alteration of “marked-to-market” values of credit default swaps by Denis Minayev, UBS staff said at an employment tribunal yesterday. Minayev, a proprietary trader, “re-marked” Braga’s trading book on 66 occasions, even though he shouldn’t have had the authority to do so, UBS investigator Richard Kennedy said.
“If you shift one of those markers, it can give a completely false picture,” employment Judge Graeme Hodgson said at the hearing.
Braga, who is suing for unfair dismissal, was an inexperienced trader who was “thrown in at the deep end,” his lawyer, Amy Sander, said at the hearing. He wasn’t aware of many of the changes Minayev made, she said, and thought his actions were permitted by managers. Braga was also accused by UBS of “procuring a false broker quote,” she said.
UBS is already dealing with the fallout from what the bank said were unauthorized trades by London-based UBS employee Kweku Adoboli, which led to a $2.3 billion loss, regulatory probes and the resignation of Chief Executive Officer Oswald Gruebel. JPMorgan Chase & Co. (JPM) CEO Jamie Dimon said yesterday his New York-based firm suffered a $2 billion loss after a trading unit’s “egregious” failure to manage risks.
Dominik Von Arx, a spokesman for Zurich-based UBS, said Braga “was a junior employee” in the bank’s fixed-income, currency and commodities unit.
“He was dismissed for gross misconduct in October 2011 following an investigation into alleged mismarking,” Von Arx said in an e-mailed statement. “UBS has zero tolerance for such behavior.”
During cross-examination of Braga today, UBS lawyer Bruce Carr said Braga had asked a broker friend to send him a quote that justified changes made to his valuation, after a colleague said the price was too high.
“You get an entirely unsolicited e-mail that happens to fit” the valuation, Carr said. “That’s quite a coincidence, isn’t it?”
Braga responded that his “dismissal shouldn’t be based on speculation or coincidences.”
The product being re-marked was a credit default swap on European industrial-company bonds, which was illiquid and difficult to value because it was rarely traded.
Lawyers for Braga questioned Paolo Croce, UBS’s European head of rates, at the continuation of the hearing today about the close relationship between proprietary traders such as Minayev, who trade with the bank’s money, and flow traders like Braga, who execute orders on behalf of clients.
“All the other flow traders followed the direction of Mr. Minayev,” Braga’s lawyer said.
Croce said while flow and proprietary traders exchanged information, they weren’t supposed to take instructions on pricing.
Minayev had told Braga, “I need this to move,” according to Croce. “He told him ‘I’m down $9 million today.’”
Here are some preliminary question to set prosecutors on their marry way?
- How many times did JPM's CIO office "procure a false quote"?
- How many times did Iksil tell his middle office or subordinates: "I need this to move" - and if he kept it to himself, how many times did Iksil "make it move" on his own?
- How many times did the CIO "shift the IG-9 market and give a completely false picture?"
- How many times did Iksil get an external "quote" that overrode the official closing day MarkIt price, or, far worse, did JPM ever tell partially-owned MarkIt what mark to use for a given product, which would be an act of unprecedented illegality.
And this is just the beginning. The reality is that with this revelation it likely means that JPM is probably lying about the fair value of thousands if not millions of other OTC-based products. Which goes to one simple thing:
Non-existent internal controls!
Because while JPM can blame an entire prop trading office for a pair trade gone wrong, it will have a very tough time explaining how marks impacting billions in P&L could have sneaked past the middle and back office.
Which, however, is possible, at least in theory.
This brings us to MarkIt - a company that has long been in the public eye for being the primary source of CDS marks, which would be great if not for one small glitch: it is also partially owned by the same banks which stand to benefit if MarkIt "nudges" the market in one way or another.
The following report from Mark Mitchell from 2009 does a great job of exposing some of the potential dirt that MarkIt may be involved in, and raises some critical questions that have never been answered, and which if addressed in the past could have spared JPM shareholders, and potentially US taxpayers, billions in losses:
Last year, the media reported that New York Attorney General Andrew Cuomo had sent subpoenas to Markit Group as part of an investigation into possible manipulation of credit default swap prices by short sellers. This investigation, like Mr. Cuomo’s other investigations into market manipulation, have yielded no prosecutions.
The Department of Justice is reportedly investigating Markit Group for anti-trust violations. This investigation (which is reportedly focused on how Markit Group packages and sells its information) seems to acknowledge that Market Group has near-monopolistic control of information about credit default swap prices. However, if the press reports are correct, the DOJ has not considered the possible appeal of this monopolistic control to market manipulators.
Meanwhile, Henry Hu, the director of the Securities and Exchange Commission’s division of risk, has said that it has been nearly impossible for the SEC to conduct investigations into any matter concerning credit default swaps because the commission does not have access to any data on the trading of CDSs. In itself, this is a shocking admission. It is all the more shocking when one considers that the necessary data exists and might be in the hands of The Markit Group – a black box company based in London.
A thorough investigation of Markit Group is urgently required.
Here is what we know so far:
- Markit Group was co-founded by Rony Grushka, Lance Uggla, and Kevin Gould. Prior to founding Markit Group, Mr. Grushka’s main line of business was investing in Bulgarian property developments. He recently resigned from the board of Orchid Developments Group, an Israeli-invested company based in Sophia, Bulgaria. Messrs. Uggla and Gould formerly worked for Toronto-Dominion Bank in Canada.
- Markit Group’s founders also include four hedge funds. However, Markit Group refuses to disclose the names of those hedge funds. In response to an inquiry, a Markit Group spokesman said it was “corporate policy” to keep the names of the hedge funds secret, but he would not say why Markit Group had such a policy. It seems worth knowing whether those hedge funds have any influence over Markit Group’s published information or indexes, and whether those hedge funds are trading on that information. It would also be worth knowing whether those hedge funds or affiliated hedge funds have engaged in short selling of public companies whose debt and stock prices were profoundly affected by the information that Markit Group published.
- Goldman Sachs (NYSE:GS), JP Morgan Chase (NYSE:JPM) and several other investment banks also have ownership stakes in Markit Group. The investment banks received their stakes in exchange for providing trading data to Markit Group. It would be worth knowing whether these investment banks engaged in short selling ahead of Markit Group’s published indexes and price quotations.
- Markit Group is secretive about how it creates its indexes. In early 2008, The Wall Street Journal noted that the CMBX simply “doesn’t make sense” and that Markit Group’s indexes “might be exaggerating the amount of distress” in the home and commercial mortgage markets. In 2008, the average prediction for defaults on commercial mortgages was 2%. The CMBX implied that the default rate could be four times that level.
- When short seller David Einhorn initiated his famous public attack on Lehman Brothers, one of his central arguments was that the CMBX (the index that was likely “exaggerating the amount of distress”) proved that Lehman had overvalued the commercial mortgages on its books.
- In March 2008, the Commercial Mortgage Securities Association sent a letter to Markit Group asking it disclose basic information about how the CMBX index is created and its daily trading volume. “The volatility in the CMBX index, caused by short sellers, distorts the true picture of the value of commercial-mortgage-backed securities,” the group said in a statement.
- Markit Group is equally secretive about how it derives its “prices” for credit default swaps. A spokesman for the company spent close to one hour talking to Deep Capture. He did his job well and sounded like he was trying to be helpful. But he told us as little as possible.
- However, in the course of this conversation, we did learn that Markit Group’s “prices” are not actual, traded prices. They are mere quotations. The Markit Group has what it calls “contributors” – hedge funds and broker-dealers that provide it with information. Markit Group has a grand total of 22 “contributors.” Deep Capture asked Markit Group’s spokesman for the names of these “contributors.” The spokesman said he would try to find out the names and call back later. He never called back.
- The 22 “contributors” provide Markit Group with quotations, and these quotations become the Markit Group’s “price.” In other words, the “contributors” can quote any price for a CDS that they choose, regardless of whether anyone is actually willing to buy the CDS at that price. Markit Group looks at these quotations. Then it somehow decides which quotations make the most sense. Then it publishes information that purports to represent the actual market price of that CDS. This process is certainly unscientific. And it is ripe for abuse.
- Consider, for example, the Markit Group “price” for CDSs insuring the debt of company X. The Markit Group price strongly suggests that company X is going to default on its debt in the immediate future. Short sellers eagerly point to the Markit Group CDS “price” as evidence that company X is doomed. Panic ensues, and suddenly, company X really is doomed. But the fact is, nobody ever bought a company X CDS at the price quoted by Markit Group. Rather, that panic-inducing “price” was, in effect, pulled out of a hat. Who pulled it out of a hat? That is matter of immense importance. There are two possible scenarios:
- The first possible scenario is that the 22 “contributors” report their quotations in good faith. They should be sending the actual traded price, not just a quotation, but assume they are just doing what was asked of them. From these quotations, Markit Group somehow decides what the “price” should be. It is possible that this decision is based on some secret formula (which would be worrisome); or it is possible that Markit Group executives sit around a table debating what the price should be and take a shot in the dark (which would be even more worrisome); or it is possible that Markit Group deliberately chooses the most horrifying price possible in order to assist the short sellers who are affiliated with its owners (which would be a matter for the authorities).
- The second possible scenario is that Markit Group acts in good faith (if not scientifically), but one or more of the 22 “contributors” or their affiliates has an interest in seeing company X fail. If just one of those “contributors” sends in an astronomically high quotation, that could be enough. Markit Group factors the absurd quotation into its posted “price” and it suddenly becomes possible to convince the world that company X is about to default on its debt. Panic ensues, the firm’s layer of debt dries up, the stock price plunges, and perhaps the “contributor” or its affiliate make a lot of money.
- As Deep Capture understands it, CDS quotations suggested by the 22 “contributors” also help determine the movement of the CMBX and ABX indexes. The movement of these indexes did serious damage to the American economy in multiple ways. The indexes prompted write downs at most of the major banks and mortgage companies. They were ammunition for short sellers, like David Einhorn, who claimed that companies had cooked their books by not writing down to the rock bottom prices suggested by the Markit Group indexes. They helped precipitate the decline in prices of mortgage securities, and contributed mightily to the panic that spread across the markets. A lot of people made a lot of money as result of those indexes moving downward. So, it is rather important to know more about how those indexes are formulated, and if they can be driven by the same people who are making directional bets on their movements.
Conclusion: Ten years ago, there was no such thing as a credit default swap. Six years ago, a very small number of investors traded credit default swaps as hedges against the long-shot possibility of corporate defaults. Nobody looked to credit default swaps as reliable indicators of corporate well-being.
Then, suddenly, there were over $60 trillion in credit default swaps outstanding. That is, over the course of a few years, somebody had made over $60 trillion (many times the gross domestic product) in long shot bets that borrowers would default on their debt. As this derivative risk marbled through the system, the trading in credit default swaps was completely opaque. Nobody knew who bought them, who sold them, or at what price.
These “prices” were not prices in any meaningful sense of the term. But, suddenly, these “prices” became perhaps the single most important indicator of corporate well-being. Assuming that those four hedge funds and the 22 “contributors” (or hedge funds affiliated with them) bet against public companies, it seems more than possible that short-sellers got to run the craps table, call the dice, and place bets, all at the same time.
So perhaps it is not surprising that a lot of long-shot rolls paid off quite nicely.
Bottom line: Jamie Dimon's "tempest in a teapot" just became a fully-formed, perfect storm which suddenly threatens his very position, and could potentially lead to billions more in losses for his firm.