The Senate Permanent Subcommittee on Investigations recently produced a 301-page report on JPMorgan Chase's (JPM 1.38%) "London Whale Trade" fiasco. The Committee's findings should be extremely troubling to investors, politicians, regulators, and concerned citizens alike.
Quoting directly from the report, here's what the investigation revealed specifically:
Over the first three months of 2012, JPMorgan's Chief Investment Office,
- used its Synthetic Credit Portfolio (SCP) to engage in high risk derivatives trading;
- mismarked the SCP book to hide hundreds of millions of dollars of losses;
- disregarded multiple internal indicators of increasing risk;
- manipulated models;
- dodged Office of the Comptroller of the Currency (OCC) oversight;
- and misinformed investors, regulators, and the public about the nature of its risky derivatives trading.
In conclusion, the Subcommittee believes its investigation:
...exposed not only the high risk activities and troubling misconduct at JPMorgan Chase, but also broader, systemic problems related to the valuation, risk analysis, disclosure, and oversight of synthetic credit derivatives held by U.S. financial institutions.
This feels like a pretty damning portrayal of a bank whose balance sheet, according to a recent research report, is "almost one-ninth the size of the United States economy." The size of just the Chief Investment Office Portfolio alone -- which was the primary subject of this investigation -- would make it the 7th largest bank in the country.
We believe the Subcommittee's painstaking investigation warrants further consideration. Below, we identify some of the most outrageous findings from the report, organized by category.
Because banker compensation is biased toward encouraging risk-taking, banks rely heavily on independent lines of risk officers and their models to keep traders in line. JPMorgan's greatest strength is supposed to be that it has superior risk management to Bank of America (BAC 0.76%) and Citigroup (C 1.17%).
Here are some risk management failures that caught our attention:
1. From Jan. 1, through April 30, 2012, Chief Investment Office (CIO) risk limits and advisories were breached more than 330 times.
2. JPMorgan's Chief Risk Officer wasn't aware of the risk problems that had already cost the bank $719 million until he read about them in the newspapers.
3. Even though the CIO was a $350 billion unit, bank managers kept giving different answers when the Subcommittee asked who its Chief Risk Officer was in late 2011. It turned out that the position was vacant in late 2011.
4. The CIO risk committee only met three times in 2011 and held its first 2012 meeting at the end of March. According to the JPMorgan Task Force Report: "There was no official membership or charter for the CIO Risk Committee."
5. According to the Subcommittee, "in 2011 and early 2012, risk managers played no role in evaluating and approving trading strategies."
6. In January 2012, the CIO settled on an unusual strategy for reducing its risk-weighted assets. Instead of simply reducing risk by selling their position (which would lose money), traders also began purchasing new, risky positions. The trade lowered how they calculated risk-weighted assets and generated income that (temporarily) covered up losses on their existing position. It's hard to overstate just how dangerous this strategy is.
7. CIO head Ina Drew said she never saw the "Decision Table" that outlined the various trading options for her in January 2012. Later, she conceded she received it as an attachment to an email, but did not focus on it.
8. On Jan. 26, 2012, trader Bruno Iksil (aka The London Whale) prepared a presentation for the CIO's International Senior Management Group advocating a new trading strategy. Iksil proposed, "Go long risk on some belly tranches especially where defaults may realize" and "...turn position over to monetize volatility." Drew admitted to the Subcommittee that the presentation was unclear to her, and she could not explain what it meant. The CIO's Chief Risk Officer at the time also told the Subcommittee that the presentation seemed unclear. The OCC advised the Subcommittee that senior management should have understood the strategy before allowing billions of dollars in trades.
9. The Subcommittee learned that the bank's "risk metrics were intended to act, not as ironclad limits, but as guidelines and red flags." CEO Jamie Dimon admitted "that a breach in a risk 'limit' was intended to lead to a conversation about the situation, not to an automatic freeze or unwinding of positions."
10. But even those conversations apparently didn't happen. "As the CIO's VaR [a key risk metric] continued to climb, the documentation produced to the Subcommittee contains few emails, messages, or telephone calls asking whether the CIO's trading strategy made sense."
11. The CIO responded to risk alarms by ignoring, minimizing, and rewriting their risk models. The head of the CIO's equity and credit trading operation estimated they could get $7 billion -- more than 50% of the total risk-weighted asset reduction they needed -- by simply modifying risk models.
12. It wasn't a violation of bank policy that the analyst who was "rushed" and "under a lot of pressure" to rewrite the VaR model worked for traders instead of risk managers and had no experience creating VaR risk models.
13. The CIO must have been desperate to get its new risk model up and running. Here's what the Subcommittee reports:
A critical risk model for a portfolio containing hundreds of billions of dollars of financial instruments, operated by the man who developed the model at the behest of the portfolio manager, included flawed and untested components, and depended upon manual uploads of key trading data daily for its calculations. This untested, unautomated, error prone VaR model was nevertheless put into place at a bank renowned for its risk management.
JPMorgan's chief regulator, the Office of the Comptroller of the Currency (OCC), later called the implementation efforts "shocking" and "absolutely unacceptable." JPMorgan's current acting Chief Risk Officer agreed.
14. Overnight, the new VaR model artificially lowered calculated risk by 50%:
15. Another risk model indicated losses could total $6.3 billion -- it proved correct. But at the time, a key CIO risk manager dismissed the figure as "garbage."
16. A separate group of key risk limit breaches grew to some 270% the allowable amounts by February. Drew agreed to increase the limit: "I have no memory of this limit. In any case it need[s] to be recast with other limits. [It is] old and outdated."
17. By April 17, 2012, a JPMorgan Chase close of business email notification stated that limit had been breached by 1,074% for 71 days.
18. The CIO didn't have any limits requiring basic diversification. The Subcommittee found that "Concentration limits are such a well-known, fundamental risk tool, that their absence at the CIO is one more inexplicable risk failure."
19. The OCC told the Subcommittee that after reviewing the portfolio's activities, it was clear that the CIO traders had made trades that violated the CIO's risk limits with "aggressive positions" in a way that was "unsafe and unsound." Also, OCC said the credit trades were "risk additive" rather than "risk reducing."
20. JPMorgan finally revoked the flawed VaR model in May 2012, after the London Whale scandal broke. But just four months later it revised the model again, lowering JPMorgan's overall VaR by 20%.
Mismarking the books
In early 2012, as the CIO began experiencing a sustained series of daily losses in its Synthetic Credit Portfolio (SCP), it suddenly began to change how it valued its positions. Traders began referring to the difference between the prices reported internally and the actual mid-point prices as "lag."
The evidence below suggests a somewhat "creative" approach to security valuation.
21. On a March 9, 2012, phone call with Bruno Iksil, fellow trader Julien Grout expressed concern about how "we're lagging," predicting that the final outcome of the trading strategy would be "a big fiasco" and "big drama when, in fact, everybody should have ... seen it coming a long time ago." He cautioned: "We have until December to cover this thing. ... [W]e must be careful."
22. By March 16, 2012, Iksil made it clear that he was using more favorable valuations for his positions. In a telephone conversation with Grout, he said:
I can't keep this going...I think what he's [their supervisor, Javier Martin-Artajo] expecting is a remarking at the end of the month...I don't know where he wants to stop, but it's getting idiotic... [N]ow it's worse than before...there's nothing that can be done, absolutely, nothing can be done, there's no hope...The book continues to grow, more and more monstrous.
23. On March 23, Iksil estimated in an email that the SCP had lost $600 million using midpoint prices, but only $300 million using "best" prices. The SCP ended up reporting a $12 million loss that day.
24. For five days in March, a trader kept a spreadsheet tracking the difference between how much money traders reported they had lost using favorable valuations, and how much they would have reported had they been using normal, mid-point valuations. The conclusion: They had reported year-to-date losses of $161 million, but there was an additional $432 million that would have been reported had they been using mid-point prices. Drew said she never saw that "shadow P&L" document.
25. At the end of the first trading day after media reports on April 6, 2012, traders responsible for reporting were told to "let the losses flow."
26. An email to Drew estimated that second-quarter losses wouldn't exceed $200 million, assuming they "exclude[d] very adverse marks."
27. A week later, Drew told traders to "[S]tart getting a little bit of that mark back... so, you know, an extra basis point you can tweak at whatever it is I'm trying to show." She later explained to the Subcommittee that traders told her they were making "conservative" marks and she wanted them to be more "aggressive."
28. JPMorgan's Controller later conducted a special investigation into the mismarking. "The Controller validated the CIO's quarter-end credit derivative marks as 'consistent with industry practices' and acceptable under bank policy, and offered no criticism of its valuation practices."
Our colleague Morgan Housel recently wrote, "Companies that have antagonistic relationships with their regulators probably want to engage in behavior that won't benefit their long-term shareholders. Bear Stearns and Lehman Brothers fought hard for permission to use more leverage. It killed them."
Here's how JPMorgan thwarted its main regulator, the Office of the Comptroller of the Currency (OCC):
29. The report found that "Prior to media reports of the whale trades in April 2012, JPMorgan provided almost no information about the CIO's Synthetic Credit Portfolio to its primary regulator, the OCC."
30. In December 2010, the OCC sent a Supervisory letter to Ina Drew informing her that "risk management framework for the investment portfolios" lacked "a documented methodology" and "clear records of decisions" and requiring CIO management to "document investment policies and portfolio decisions." Drew "sternly" criticized the OCC for 45 minutes for being overly intrusive. She complained the regulator was trying to "destroy" JPMorgan by taking away necessary flexibility, and said that Jamie Dimon was fully aware of their investment decisions.
31. According to the OCC examiner-in-charge at JPMorgan, it was "very common" for the bank to resist OCC findings and recommendations. One time bank executives yelled at OCC examiners and called them "stupid." Another time, the most junior-level OCC examiner arrived for a one-on-one meeting and was "ambushed" by the heads of all JPMorgan Risk Divisions.
32. In January 2012, the CIO seems to have misled the OCC by saying it planned to reduce the Synthetic Credit Portfolio. Instead, it tripled the notional size of the SCP from $51 billion to $157 billion over the course of the quarter.
33. As losses grew from January through April 2012, the bank began providing the OCC less and less data.
34. Dimon ordered the bank to omit critical CIO performance data from its standard reports to the OCC. No one told the OCC why the data was halted. Dimon later explained to the Subcommittee that he didn't feel the OCC needed the data.
35. Dimon "raised his voice in anger" at CFO Douglas Braunstein for agreeing to resume sending the OCC the reports.
36. When the OCC found out about the whale trade in the newspapers and requested more information, the CIO gave them incomplete tables that the OCC called "useless" and "absolutely unhelpful."
37. According to the Subcommittee, "For ten days, from April 9 to April 19, the bank repeatedly assured the OCC that the CIO whale trades were nothing to worry about." It told the OCC the whale trades helped manage bank risk and gave different excuses for breaking risk limits. Then, "for nearly three weeks, the bank did not call, email, or otherwise update the OCC about any aspect of the SCP's worsening status." By that point, losses had reached $1.6 billion.
According to a yet-to-be-implemented piece of the 2010 financial reform legislation known as the "Volcker Rule," banks are supposed to be prohibited from speculating with their funds. While banks have managed to delay the rule's implementation, in the meantime, they've been keen to present themselves to investors, regulators, and the public as out of the speculation business.
Although banks will be able to continue placing hedging trades to help manage risk -- and that's what JPM asserted it was doing -- that doesn't appear to be what happened here.
38. The original derivatives positions came from a discontinued unit called "Proprietary Positions Book," and the portfolio used to be part of something called the "Discretionary Trading Book" that was renamed the "Tactical Asset Allocation Portfolio" -- jargon that suggests speculation.
39. The SCP was supposedly created with a hedging function, but the bank was unable to provide documentation over the next five years detailing SCP's hedging objectives and strategies; the assets, portfolio, risks, or tail events it was supposed to hedge. One OCC examiner referred to it as the "make believe voodoo magic 'composite hedge.'" Even Dimon acknowledged that the portfolio "morphed into something that rather than protect the firm, created new and potentially larger risks."
40. JPMorgan representatives admitted to the Subcommittee that calculating the size and nature of the hedge was "not that scientific" and "not linear." According to Drew, it was a "guesstimate."
41. Compensation history for SCP traders shows that the bank rewarded them for financial gain and risk-taking more than for effective risk management.
42. In late 2011, the SCP bankrolled a $1 billion bet that produced a gain of $400 million. This was known as the caveman trade. Iksil said the gains were massive -- and the company called them "windfall gains." The OCC would later characterize the trade as "pretty risky" and completely dependent on timing. Drew responded by telling her traders to try and repeat their performance the following year.
43. In 2011, the SCP net notional size jumped from $4 billion to $51 billion, a more than tenfold increase. In the first quarter of 2012, the net notional size increased to $157 billion. By end of March 2012, the SCP held more than 100 different credit derivative instruments -- JPM personnel described it as "huge" and of "perilous size."
Once the London Whale trades became public, JPMorgan's executives worked to reassure investors, markets, and the public that the whale trades were a hedge (as opposed to speculation) and that the bank had things under control. The Subcommittee was skeptical of how truthful JPMorgan's management has been with investors and the public.
44. During JPMorgan's April 13 earnings call and its May 10 "business update" call, CEO Jamie Dimon and CFO Douglas Braunstein reassured investors and the public that the portfolio was a long-term hedge for helping the bank manage risk. According to the report, the two executives struggled to explain to the Subcommittee why they believed the portfolio functioned as a hedge.
45. On April 13, Braunstein also told investors that the bank believed its SCP trades were consistent with the approaching ban on bank speculation. The bank was unable to provide the Subcommittee with any legal analysis supporting that claim. In fact, the bank's official public comment letter to regulators said just the opposite of what the bank told investors.
46. Braunstein reassured investors that "all of these positions are put on pursuant to risk management at the firmwide level." According to the Subcommittee, "virtually no evidence indicates that he, his predecessor, or any other firmwide risk manager played a role in designing, analyzing, or approving the ... positions acquired in 2012." In fact, JPMorgan's Chief Risk Officer didn't find out about the SCP's problems until they were reported in the press. Braunstein also didn't mention that the CIO had blown through its risk limits.
47. Braunstein also told investors, "[A]ll of those positions are fully transparent to the regulators. They review them, have access to them at any point in time, get the information on those positions on a regular and recurring basis as part of our normalized reporting." According to the Subcommittee, "This statement by Mr. Braunstein has no basis in fact." The OCC thought the bank was phasing out the SCP portfolio and wouldn't even be told about its losses for another three weeks.
48. The Subcommittee points out that in an April 13 public filing with the SEC, JPMorgan indicated to investors that the CIO's risk had slightly declined, even though the SCP portfolio was three times larger, had riskier assets, and the bank had replaced its previous risk model with a new one that dramatically lowered how risk was calculated.
On July 13, 2012, the bank restated its first-quarter earnings, reporting additional SCP losses of $660 million. It struggled with that, thinking the loss wasn't "material."
What's the big deal?
Some observers might be tempted to dismiss the "Whale Trade" fiasco as a relatively minor mistake at an enormously complex financial institution -- a "tempest in a teapot" if you will. The overall losses were at least $6.2 billion -- a very manageable amount for JPMorgan, which had shareholder equity of $184 billion on its balance sheet at the end of 2011.
The problem, of course, is that a future loss might be considerably larger. Just recently, JPMorgan itself examined the effects of a "hypothetical" loss of $50 billion. That exercise showed that the bank would fail in that scenario.
So ultimately, we all must ask: What is the likelihood that JPMorgan might suffer a loss that could put its survival -- and the stability of the entire financial system -- at grave risk? We believe the evidence presented above suggests that likelihood is much higher than it should be. What do you think?
If you want us to keep you up-to-speed on financial reform and investor protection, just shoot a blank email to [email protected].