NuclearPhynance Archive

Anyone have anything to say about the Report of JPMorgan Chase & Co. Management Task Force Regarding 2012 CIO Losses ?



Some of the more sensationalist comment so far at

BI - Excel-partly-to-blame-for-trading-loss which is rubbish of course it wasn't Excel per se, but its poor use and apparent lack of adequate model validation certainly did contribute



See also:Baseline - The Importance of Excel AND



FT Alphaville - A Tempest in a Spreadsheet

Thanks for posting that.  There is too much material to comment on so I'll just point to that one paragraph about Goldman being appointed chief risk officer even though he had no idea how to do that job.  I've seen that before.  These banks look around for the alpha males, ones that are good at golf and such, and annoint them risk officer as if its something they could do automatically just by trying.  Similar to being chosen to do brain surgery later that day if you really try.  There's lots of modellers that have forgotten more than he knows but would never get anywhere near that selection process.  The bank should not be surprised by the results any more than if brain surgery was done a Walmart clerk.

I just had a quick glance to fresh up my beliefs.



As far as I can see it this is a curve trade gone bad... unfortunately in size. Long 20bn CDX IG 9 10y, short 12bn CDX IG 9 5y. Yeah, I guess their models told them at that time that this is a market-neutral position under a 10% spread widening scenario. It might even have been. Unfortunately, this size leaves less room for error.



I didn't check when the series started but this might also be a relevant point. 5y means that at inception of the series time to maturity is 5 years (5.25 to be precise). Since IG 9 was around for a long time the original 5y might have become a 3y and the original 10y might have become an 8y. Just guessing, but I think you get the idea. Unfortunately your delta (or DV01 or CSW10% or [insert arbitrary spread risk measure]) becomes much less sensitive as maturity nears. So your 10y position becomes less risky as time goes by, fine, but so does your 5y position. Only to a much higher degree...



I didn't dig into the short CDX HY positions yet, seems like the had a macro hedge on the books that was financed with a CDX IG long position (at least partly since they paid out before 2011).



What comes on top, though, is that probably everyone in the street knew that JP had a very sizeable CDX IG 9 position. IG 9 was the most liquid series for a long time (10 and some series afterwards were spoiled by crap names) but even there you don't built up a 20bn position and nobody notices it.

After Huge Blowup, Senate Report Said to Fault JPMorgan Execs

The New York Times | March 05, 2013 | 01:42 AM EST



While a trader known as the "London whale" has come to represent a multibillion-dollar blowup at JPMorgan Chase, Congressional investigators have discovered that the problems involved more senior levels of the nation's largest bank.



A report by the Senate Permanent Subcommittee on Investigations highlights flaws in the bank's public disclosures and takes aim at several executives, including Douglas Braunstein, who was chief financial officer at the time of the losses, according to people briefed on the inquiry. The report's findings — scheduled to be released on March 15 — are expected to fault the executives for allowingJPMorgan [ JKMZZ 0.00 +0.00 (+0.00%) ] to build the bets without fully warning regulators and investors, these people said.



The subcommittee, led by Senator Carl Levin, could ask Mr. Braunstein and other senior executives to testify at a hearing this month, according to the people. The subcommittee does not currently intend to call the bank's chief executive, Jamie Dimon, but Congressional investigators interviewed Mr. Dimon last year.



JPMorgan, which has been cooperating with the investigation and discussed the findings with the subcommittee, declined to comment. Mr. Braunstein and other bank executives have not been accused of any wrongdoing, and he is not the focus of a separate law enforcement investigation into the trading loss.



Congressional officials have yet to set the final details of the hearing and plans may change, the people cautioned. Politico earlier reported the scheduled date for the release of the report.



The Congressional investigation could revive questions about the role of senior executives in the $6 billion trading loss at a time when the bank has started to put the blunder behind it.



Mr. Dimon declared last year that the "Whale has been harpooned." The bank reported record earnings in January and has forced out the architects of the bet.



The Senate report, however, shifts the focus from lower-level traders in London who placed the bet to senior executives and regulators who failed to stop it. Expanding on a sweeping report the bank released in January, the Congressional inquiry is expected to open a window into how executives ignored warning signs and failed to alert investors about changes to its method for detecting risk.



Because a large majority of the executives involved in the trade have since departed the bank, the hearing could increase scrutiny of Mr. Braunstein and Mr. Dimon, the remaining senior executives. Within JPMorgan, people close to the bank say, executives have expressed dismay about the lingering questions.



The report, a reminder that Wall Street blowups continue even four years after the financial crisis, could galvanize support for regulations like the Volcker Rule that aim to rein in risky trading. Mr. Levin, a Democrat of Michigan who champions the Volcker Rule, is expected to use the report to endorse policy changes, including stricter public disclosures.



But Mr. Levin's staff is still negotiating with the committee's Republicans over the recommendations. John McCain, the ranking Republican, has largely approved the report's findings but continues to examine the policy ideas, the people said.



A spokesman for Mr. McCain declined to comment.



The subcommittee's report coincides with a federal investigation into four employees in London, including Bruno Iksil, the so-called Whale, who carried out the trades at the bank's chief investment office. The Federal Bureau of Investigation is conducting inquiries into some of the traders, according to officials, suspecting they hid problems from the bank.



But the subcommittee's investigators seized on e-mails suggesting that Mr. Iksil had raised alarms about the bet. In an e-mail to a more senior trader in January 2012, he advised against increasing the bet, according to people who reviewed the message. The size of the trades, Mr. Iksil said, were becoming "scary" and advised that the investment office take the "full pain" now, according to a person briefed on the e-mails. JPMorgan released the e-mails without naming the traders.



By February, Mr. Iksil grew worried as he struggled to understand why losses were escalating. Later that month, he instructed a junior trader to temporarily halt trading. Their boss later reversed that decision.



The subcommittee's report is expected to detail how senior executives failed to heed warnings from London. Some of those findings echo JPMorgan's report, released this January, which examined the role of Mr. Braunstein; Ina Drew, who led the chief investment office; and Barry L. Zubrow, a former chief risk officer. Ms. Drew and Mr. Zubrow have since left the bank.



Scrutiny around Mr. Braunstein, who is now a vice chairman at the bank, partly focused on his reliance on other people's assurances about the safety of the trades. In its own analysis of the trade, JPMorgan said Mr. Braunstein incorrectly assumed that the positions in the chief investment office were "manageable."



The focus on Mr. Braunstein also stems from the bank's inconsistent statements. He dismissed concerns about the positions in April 2012, assuring analysts in a conference call that the bank was "very comfortable with our positions." The subcommittee has examined whether those disclosures were misleading.



The subcommittee further examined whether the bank failed to alert investors about a change in its internal alarm system. The bank in January 2012 introduced a new value-at-risk model that underestimated the losses in the investment office. The bank did not inform investors about the model change until May.



In the lead-up to the subcommittee's reports, the bank faced questions about similar disclosures to regulators. In some instances, the people briefed on the report said, bank employees initially resisted requests from regulators at the Office of the Comptroller of the Currency who sought deeper details.



But regulators will not escape criticism in the report.



The bank warned some regulators about the changing risk model, a person briefed on the matter said. In an e-mail to an official in the comptroller's office, the bank disclosed that the new model could cut its risk in half, something that might have been viewed as a startling revelation.

(1) Definitely, the firm’s Chief Investment Office needs to create a standardized methodology for financial software development in Excel.



For a good Excel development methodology, refer to the paper “A Software Development Methodology for Research and Prototyping in Financial Markets” by Andrew Kumiega, Ph.D. and Ben Van Vliet. http://arxiv.org/pdf/0803.0162.pdf



As a technical reference see the large collection of Mathematical and Statistics libraries that can be used in documentation of Excel based algorithms and financial calculations: https://github.com/rnfermincota/IVEY/tree/master/modules



(2) Use Hash Tables in Excel to get and trace data from multiple worksheets, workbooks and/or text files.



For an effective implementation of a Hash Table in Excel see the Stock Market add-in of the Ben Graham Centre for Value Investing. https://github.com/rnfermincota/BGCVI



(3) Last but not least, banks need to consider using R Programming Language for Risk Management and Trading Applications. Check the excellent work done by Dirk Eddelbuettel: http://gallery.rcpp.org/ http://cran.r-project.org/web/views/Finance.html

Bankers Trust was pretty good at cobbling together an excel system for exotics...a 1000 years ago.

Of course Bankers Trust Exotics kicked ass at all times...saving the bacon on several occasions.

I think the problem there was it was primarily their clients asses that were kicked - Perhaps that's were the recent and current trend started.

True but it was a pretty coo place.

Full text of Senate Report on JPM CIO Office available here