JP Morgan Chase Whale Trades: A Case History of Derivatives Risks and Abuses

JP Morgan Chase Whale Trades: A Case History of Derivatives Risks and Abuses

March 19, 2013

On March 15, 2013, the Senate’s Permanent Subcommittee on Investigations held hearings on the London Whale scandal. The indomitable and indefatigable Chairman Carl Levin, ably supported by the brilliant committee chief of staff, Elise Bean, took on six JP Morgan Chase (“JPMC”) current and former executives for four hours and three regulators for two, with support from other Committee members. The Senator used the information from the 300-page committee report masterfully to illicit information on the massive losses incurred by JPMC’s London branch on a trading strategy involving $157 billion notional amount of credit derivatives tied to the default risk of corporations in Europe and the US.

This hearing and the accompanying report reveal specific misdeeds and the chilling fallibility of risk systems that could plunge the word into economic ruin. They also provide insights into Wall Street culture and its operational realities. The consequences of the hearing and report are potentially enormous, changing public discourse on controlling the mega-banks, who see themselves as a law unto themselves, governed by greed and hubris worthy of Greek tragedy. The London Whale trades could not have brought JPMC down; the losses, $6.2 billion, were too small given the size of JPMC. But the risky behavior, incompetence, lawlessness and arrogance very well could some day.

There are several broad areas of misconduct covered by the PSI: disclosure and regulatory reporting are two. These grow out of the other broad area: the preference for manipulation of analytics to obscure reality losses and risks growing out of unrestrained trading behavior. The following is a synopsis of these topics together with an analysis of the implications of the hearing and report.

Manipulation of Analytics

By the end of 2011, the department of JPMC known as the Chief Investment Office (“CIO”), whose task was to reduce risk in the global organization, held a portfolio of indexed credit default swaps with a notional amount of $51 billion (the “Synthetic Credit Portfolio” or “SCP”). Consistent with CIO’s mission, the SCP was denoted a “hedge,” or offset to risks of JPMC’s business. The business risks hedged were obscure, however. They were described as the risk of companies around the world going bankrupt. The SCP had, in reality, become a complex set of transactions that seemed not to hedge anything in particular. Risk mitigation had increasingly become secondary to trading for profit over a three-year period in which the SCP had grown ten-fold in notional amount. The witnesses confirmed that, regardless of the initial intent, the SCP was a massive risk position designed to make (or lose) large sums of money.

This very fact was in fact a manipulative deception. The CIO was designed to invest excess deposits (i.e., deposited funds that had not been loaned) in safe securities and to enter into hedges that would offset bank risks. This deflected the attention of regulators from the CIO who ascribed a low priority to red flags associated with the CIO that could have attracted inquiry. Based on appearances, there were other areas of the bank that were far more susceptible to dangerous losses.

By early 2012, the SCP had created two separate problems for the CIO. First, it simply started to yield losses rather than profits and the losses were large. Second, the internal risk management calculation of forward-looking risk meant that, under the soon-to-be implemented international Basel 2.5 rules, the SCP would force JPMC to set aside large amounts of capital to offset its risks. This was the result under Basel 2.5’s risk weighted capital rules, or “RWA.” Instead of winding down the SCP, the CIO and other parts of the bank set out to solve these two problems in extremely dubious ways.

The cumulative market gains or losses were determined by comparing current market prices of the SCP positions with their acquisition costs. This is referred to as “marking to market.” Market prices are really the prices to buy and sell that are quoted on the markets. The spread between the two is the “bid/ask” spread. JPMC, and most financial institutions, had commonly used the midpoint between the bid and ask prices to mark their holdings to market.

In January, the CIO traders abandoned that practice and started using prices closer to bid or ask, the boundaries of the bid/ask spread. Uniformly, they picked prices at the border of the bid/ask spread that lowered the measured losses in the SCP compared with marks based on the midpoint prices. The effect on the understatement of losses was not cumulative. For example, assume a position acquired a 7. If the current midpoint price is 5 and the manipulated price is 6 on day one, the mark would be the difference between 7 and 6 rather than 7 and the midpoint, or 5. The misstatement of cumulative loss is 1. On the next day, if the midpoint price is 4 and the manipulated price is 5, the cumulative loss would be measured at 2 rather than 3, still a cumulative misstatement of 1. Of course the cumulative loss would be misstated every day, but never more than the current difference between the midpoint and the highest price within the bid/ask spread. It was a manipulation that did not get substantially worse quantitatively each day, but was persistent over a period of months.

As has become familiar in investigations of financial institutions, e-mails told the story. The CIO traders who applied the marks called them “idiotic” and openly discussed what they were doing. Later in the first quarter, senior bank management defended the marks internally as within GAAP accounting rules, which is of course beside the point. Senator Levin successfully forced the bank witnesses to agree that manipulating mark-to-market practices for the purpose of hiding losses in a portfolio is unacceptable, even if it were in accordance with GAAP, an assumption that he did not agree with.

In bilateral derivatives transactions, margin collateral is often posted based on marks that are to be agreed to by the two counterparties. That was the case with the SCP. The marks used by the CIO traders for SCP precipitated serious disputes with counterparties starting in March. The amounts disputed peaked in mid-April at $690 million. These disputes, which were a clear indication of problems with marks, came to the attention of senior bank officials and the regulators in April, after news accounts had revealed the issues associated with the SCP.

Notwithstanding losses as early as January 2012, the CIO doubled and tripled down on the SCP positions. The SCP grew to $157 billion in notional amount by the end of the first quarter, even though the CIS had told regulators in January that it intended to reduce its size over 2012. The traders were literally out of control. As the SCP got bigger and the indexed credit derivatives market was showing signs of stress, in large part because of CIS’s buying spree, the forward-looking measurement of risk associated with SCP deteriorated badly. This set off alarms within JPMC, because risk limits were being exceeded and the RWA Basel 2.5 calculations were requiring more capital to be taken out of circulation.

The primary (but not exclusive) problem for CIO was two variations of Value at Risk, a statistical estimate of how much the bank might lose in SCP in one day of trading based on historic experience. One variation looked at “normal” market conditions and the other looked at “stressed” market conditions. Limits were exceeded for both (as well as for other, less critical metrics).

JPMC addressed the breeches of VaR limits by temporarily increasing the limits, an act that CEO Jamie Dimon approved. This was intended as a stopgap until the VaR mode could be changed. The CIO had been constructing a new VaR model that would lower calculated risk in the SCP. It had not yet been properly tested and there was no system in place to transmit data on new trades automatically. New trades would have to be manually recorded on a spreadsheet and then loaded each night, a cumbersome process fraught with the potential for human error.

Nonetheless, CIO put the new VaR model in place at the end of January. The apparent risk of the SCP dropped by 50% instantaneously, well below limits. Nothing had changed, but the flashing red lights were turned off. And the original risk limits were reinstated, not new more stringent limits appropriate to the new model’s built-in lower risk calculations.

It would be reasonable to anticipate that CIO would cut its losses and reduce the SCP in an orderly fashion. But instead, the traders used the new risk headroom to lay on more trades, increasing the SCP notional amount to $157billion by the time the “phones down” order came from management at the end of the first quarter stopping all trading in the SCP. In the interim, the manual data inputs had, predictably, created a corrupted data set that added to the problems of the new model. The old VaR model was reinstated in May, after press reports had made the London Whale trades public and billions had been lost.

Misleading Regulators

JPMC’s principal prudential regulator, the Office of the Comptroller of the Currency (“OCC”) first learned of the massive SCP position from the initial press reports of the London Whale on April 6. This is particularly concerning given the size of the risk associated with the SCP and the fact that the OCC has 65 onsite examiners dedicated to JPMC.

Within days of becoming aware of the SCP through the press, the OCC determined that the SCP was not actually a hedge, but was in fact a massive risk position designed with the intent to generate trading profits. After May 4, the OCC met with JPMC daily regarding the SCP. But, by then the bank had already lost at least $2.3 billion on the positions. (Note that the OCC claims that CFO Douglas Braunstein said on May 4 that the losses were $1.6 billion, but Braunstein claims that the $1.6 billion number was identified as the losses since the start of the 2nd quarter on April 1.)

Prior to that time, the OCC was aware that the CIO had breached the risk management limits according to several criteria. Of course, the VaR related criteria were changed to eliminate critical breaches, but breaches under other, perhaps less critical, criteria persisted. A clear opportunity to influence the outcome had been lost and the SCP grew immensely after the initial breaches.

The OCC was unaware of the changed mark-to-market procedures until May when it was informed of the counterparty collateral disputes and enquired further.

The OCC regulators explain that they failed to pursue warning signs promptly largely because the activities of the CIO were thought to be a low priority in terms of potential risk. They had no idea of the size of the SCP. And the OCC assumed that the CIO activities were in accord with its stated mission, investment of excess deposits and risk-reducing hedging. This does not explain why the OCC was so unaware of CIO activity.

The OCC asserts that JPMC, like all large banks, is expected to affirmatively inform the OCC of significant events, such as the size of the SCP, the losses, the occurrence of the risk limit breaches and the changes to the VaR models. That did not occur. This failure is consistent with a pattern of activity at JPMC.

The PSI report and hearing depict a very different relationship between JPMC and its principal regulator than that described by OCC as appropriate. In 2010, the OCC issued a Supervisory Letter directing improvement of CIO’s portfolio decisions and risk management. The head of CIO and Operating Committee member, Ina Drew, strenuously resisted the admonition in a meeting, claiming that the OCC was trying to “destroy” JPMC’s business and reportedly asserting that “that investment decisions are made with the full understanding of executive management, including Jamie Dimon.”

Another example put forth by PSI involved the abrupt and unexplained cessation of daily P&L reports concerning JPMC’s investment bank in February 2012. Testimony states that CEO Jamie Dimon ordered this cessation because he believed P&L information had leaked, though there was no evidence that the OCC was the source of the leak. The OCC eventually complained to the then-CFO Douglas Braunstein who reinstated them weeks after the cessation, an act that angered of Mr. DImon. Senator McCain remarked on Jamie Dimon’s decision, musing whether there was any other company in America that would, in an unfounded fit of pique, stiff its principal regulator on required reports and expect to get away with it.

Other antagonistic meetings were recounted. What emerges is an adversarial attitude of JPMC executives toward the OCC resulting in an adversarial relationship that is simply inappropriate.

It must be pointed out that the current Comptroller of the Currency was not confirmed and sworn in until April 2012, replacing an Acting Comptroller who had been in charge of OCC for more than 1 ½ years. Curry came to the OCC amidst reports of the agencies longstanding lax attitude toward bank regulation. The hearing and report indicate that the Commissioner needs to make changes.

Misleading the Public

JPMC disclosed nothing publicly about SCP prior to news stories on April 6. Clearly, widespread knowledge of the situation could have made management or liquidation of the SCP more difficult.

As stated in the report, on April 10, as internal estimates of the loss had reached $415 million, “[t]he bank’s communications officer and chief investment liaison circulated talking points and, that same day, April 10, met with reporters and analysts to deliver reassuring messages about the SCP. Their primary objectives were to communicate, among other matters, that the CIO’s activities were ‘for hedging purposes’ and that the regulators were ‘fully aware,’ of its activities, neither of which was true.”

JPMC followed these messages up on April 13 by filing an 8-Q report with the SEC and hosting an earnings call featuring CEO Jamie Dimon and then-CFO Douglas Braunstein. DImon referred to the media reports as a “tempest in a teapot.” Braunstein made several remarks that the PSI considered concerning, as demonstrated by the following quoted text from the report:

“Mr. Braunstein stated on the April 13 earnings call that ‘all of those positions are put on pursuant to the risk management at the firm-wide level.’ The evidence indicates, however, that in 2012, JPMorgan Chase’s firmwide risk managers knew little about the SCP and had no role in putting on its positions.”

  • “Mr. Braunstein said that the SCP positions were ‘fully transparent to the regulators,’ who ‘get information on those positions on a regular and recurring basis as part of our normalized reporting.’ In fact, the SCP positions had never been disclosed to the OCC in any regular bank report. The bank had described the SCP’s positions to the OCC for the first time, in a general way, only a few days earlier and failed to provide more detailed information for more than a month.”
  • “Mr. Braunstein also stated that with regard to ‘managing’ the stress loss positions of the Synthetic Credit Portfolio, ‘[a]ll of the decisions are made on a very long-term basis.’ In fact, the CIO credit traders engaged in daily derivatives trading, and the bank conceded the SCP was ‘actively traded.’ An internal CIO presentation in March 2012, provided to the bank’s executive committee a month before the earnings call, indicated that the SCP operated on a ‘short’ time horizon. In addition, many of the positions producing SCP losses had been acquired just weeks or months earlier. Mr. Braunstein’s characterization of the SCP as making long term investment decisions was contrary to both the short-term posture of the SCP, as well as how it actually operated in 2011 and 2012. His description was inaccurate at best and deceptive at worst.
  • “Mr. Braunstein indicated that the SCP was intended to provide ‘stress loss protection’ to the bank in the event of a credit crisis, essentially presenting the SCP as a portfolio designed to lower rather than increase bank risk. But in early April, days before the earnings call, Ms. Drew told the bank’s executive committee that, overall, the SCP was ‘long’ credit, a posture that multiple senior executives told the Subcommittee was inconsistent with providing protection against a credit crisis.”
  • “[Mr. Braunstein said:] ‘[W]e believe all of this is consistent with what we believe the ultimate outcome will be related to Volcker.’ The Volcker Rule is intended to reduce bank risk by prohibiting high risk proprietary trading activities by federally insured banks, their affiliates, and subsidiaries. However, the Volcker Rule also allows certain trading activities to continue, including ‘risk-mitigating hedging activities.’ Mr. Braunstein’s statement gave the misimpression that the SCP was ‘hedging’ risk.”

The PSI also expressed concerns about the omission of disclosure about changes to the VaR model until after the old VaR model was reinstated. This was particularly concerning because the April 13 8-Q reported a first quarter CIO VaR figure that was comparable to the prior reported CIO VaR figure for the last quarter in 2011 without stating that the VaR model had been changed in the interim. The PSI was also concerned with the characterizations of the reasons for reinstating the old VaR model when it was discussed in a May 10 call with investors and analysts in connection with the filing of the banks annual 10-K report with the SEC.

Implications of the PSI Hearing and Report

The issues that have been raised have many implications for policies and regulations that are currently being deliberated in the United States and abroad.

The Volcker Rule

The Volcker Rule generally prohibits insured banks from engaging in proprietary trading, or trading activities that put their balance sheets at risk. Proposed regulations have been issued and final regulations are being prepared.

The statute excepts the following activity from the proprietary trading ban:

Risk-mitigating hedging activities in connection with and related to individual or aggregated positions, contracts, or other holdings of a banking entity that are designed to reduce the specific risks to the banking entity in connection with and related to such positions, contracts, or other holdings.

The proposed rules contain language that appears to interpret this provision narrowly and other language that is broader. This requires reconciliation in the final rule.

JPMC asserted that the SCP was originally a hedge that morphed into a proprietary position and that the hedge would have qualified under the Volcker Rule. The risk that was “hedged” appears to have been a general amorphous risk from corporate defaults. Was this risk related to “individual or aggregated positions, contracts or other holdings?” Furthermore, was it designed to “reduce the specific risks?” Critics of the Volcker Rule assert that positions like the SCP, justified as offsetting some ill-defined risks, should be permitted. The PSI hearing and report provide persuasive evidence that the hedging exception should instead be tightly drawn in the final rules.

Moreover, the proposed rules include a series of metrics to be reported on by the banks that are used to flag possible proprietary trading activity, even if it is labeled by the bank as another type of activity that would be permitted. The purpose of the CIO was not proprietary trading, but that is what it did. The critics of the proposed rules claim that the metrics are burdensome and unnecessary. The London Whale episode demonstrates the wisdom of using the metrics as proposed.

Finally, the Volcker Rule allows an exception for “market making,” a low-risk business that provides customers access to transactions based on current market price quotes. The proposed rules stretch this concept to include customer transactions based on prices assigned by traders rather than the markets. The manipulation of prices by CIO traders illustrates the unreliability of trader marks. This activity is not market making. It is taking a proprietary risk position and should not be permitted.

Extraterritorial Jurisdiction of the CFTC

The CFTC is responsible for regulating derivatives. Under Dodd-Frank, its jurisdiction extends to activities that "have a direct and significant connection with activities in, or effect on, commerce of the United States." The application of this jurisdiction to branches and affiliates of US banks is a subject of vigorous debate and proposed legislation that would amend Dodd-Frank.

The London Whale trades were made at the London branch of JPMC. The consequences were visited on JPMC and its share price and reputation suffered accordingly. The PSI hearings and report make it abundantly clear that these trades had a direct and significant connection with, and effect on, the commerce of the United States. They demonstrate that the CFTC’s jurisdiction extends to the CIO’s London operations under Dodd-Frank. They also demonstrate that this is a good result and the proposed amendments are ill advised.

Capital and Leverage Requirements

Proposed regulations impose requirements related to bank capital and leverage, consistent with the proposed Basel international standards. The implementation of these requirements relies on the measurement of the risks associated with bank assets and other metrics that are the responsibility of the banks in the first instance.

The actions of the JPMC related to marks-to-market and VaR should inform the approach taken by the regulators to these matters. It is not enough to require that metrics be specified. How the metrics are developed and used is critically important. Marks must be based on objective criteria, requiring at least prices provided by third-party index providers, but preferably derived from actual market transactions. And risk models need to be thoroughly vetted based on sound standards. Capital and leverage requirements are only useful if the measurement processes are reliable, transparent and prudent.

Authority of Regulators

The PSI hearings and report depict management of a major bank that arrogantly disregarded its regulators and cavalierly misled them and failed basic transparency duties. The regulatory process cannot function if the regulated banks behave as righteous adversaries when dealing with their regulators. Stronger penalties for this behavior are needed to enhance the authority of the regulators.

In addition, the tasks of regulators of major international financial organizations are daunting. The complexity of these institutions is a major obstacle. This complexity needs to be curbed. But, additionally, the regulators need to be provided the resources to do their jobs. This will reinforce their credibility with the mega-banks.

The confirmation process should not be used to deny these important agencies proper leadership. And they should be provided the resources to effectively regulate the enormously complex financial sector. The resources needed by the financial regulators are modest, in the context of both the Federal budget and the scope of the financial sector. The financial crisis (and, indeed the London Whale episode) illustrates the public’s interest in well-resourced financial regulatory agencies. when dealing with their regulators. Stronger penalties for this behavior are needed to enhance the authority of the regulators.

In addition, the tasks of regulators of major international financial organizations are daunting. The complexity of these institutions is a major obstacle. This complexity needs to be curbed. But, additionally, the regulators need to be provided the resources to do their jobs. This will reinforce their credibility with the mega-banks.

The confirmation process should not be used to deny these important agencies proper leadership. And they should be provided the resources to effectively regulate the enormously complex financial sector. The resources needed by the financial regulators are modest, in the context of both the Federal budget and the scope of the financial sector. The financial crisis (and, indeed the London Whale episode) illustrates the public’s interest in well-resourced financial regulatory agencies. 

Read the Demos Storify of the London Whale Hearings