Sort by

The London Whale: Tempest in a Teapot or Cover-up?

The Senate Permanent Subcommittee on Investigations hearing and 300-page report (plus 200 pages of exhibits) on the London Whale is now being digested by the public and spun by JP Morgan Chase. The bank lost $6.2 billion and survived easily because of its immense size. Will it turn out to be a “tempest in a teapot,” the phrase that CEO Jamie Dimon used in an investor and analyst conference call one week after the press broke the story back in April 2012?

It depends on whether the press, the regulators and the Justice Department keep their eyes on the real reason the episode was important. They must avoid being distracted by the JP Morgan Chase spin machine. Like a cheap nightclub magician, the bank wants everyone to look at the loss. Their spinners want us to agree that, in the context of the vast balance sheet, the whole thing was no big deal.

But the management of the bank did not act as if it were trivial, even before the story broke in the press. That is because the importance of the London Whale was not so much what was lost. It was that the vaunted risk controls of JP Morgan Chase allowed the trades to happen.

The London Chief Investment Office was out of control. Its sole stated mission was to do two things: invest excess deposits (i.e., deposited funds that had not been loaned), many of which are insured by the Federal government in safe securities; and to enter into trades, or “hedges,” that would offset bank risks. It is obvious that the CIO was actually betting, and betting big time, on risky credit default swaps, the same type of instruments that ignited the financial crisis.

An investor in JP Morgan Chase would be far more interested in whether the bank’s risk management operation, which had been touted as the world’s best by the bank, all the way up to the CEO, could control its far-flung worldwide operations than in a trading loss. Loss of confidence in the bank’s risk management would not be a “tempest in a teapot.”

The lengths to which the bank would go to hide this problem are evident in the hearing and report. Senator Carl Levin, in his brilliant questioning of witnesses, exposed a portfolio of bad acts by the bank. One story involves the risk systems used by the bank. Here is what the Subcommittee report lays out.

In its fourth quarter 8K report filed with the SEC, JP Morgan Chase provides a figure for the “Value at Risk” associated with the CIO. Value at Risk, or VaR, is a statistical calculation of what the bank might be expected to lose in one trading day o a portfolio of investments, given historical price movements. The figure that the bank gave for the CIO in the 8K was $64 million.

By January that number was skyrocketing. It climbed to $132 million by the end of the month. The number breached internal limits applicable to CIO in mid-January on its way higher. This was a problem. It was not that the bank could not tolerate the risk as measured, it was that the breach of risk limits would call attention to the fact that JP Morgan Chase’s management had condoned the CIO’s transformation into a risk oriented trading business and then allowed it to run wild. 

The bank’s response was to change the apparent facts. Jamie Dimon approved a temporary increase of the CIOs risk limit, turning off the flashing red light occasioned by the breach. But this could not have been a real fix, being temporary after all. Management had to have something else in mind to solve their problems. It turns out that bank employees had been working on a new VaR model for the CIO and they offered up the observation that VaR would be cut by 44% if it were put online. That is to say, the risk would not change, just the apparent risk.

But it would be sensible to test the model by “back checking,” that is to say run it for prior months and compare the results with those generated by the existing VaR model. In addition, no automatic feed of trade data had been created. To run the new VaR model the trades for each day would have to be input by hand every night. On January 26, the new VaR model went live anyway. Sure enough, the VaR figure dropped by 50% instantaneously. Of course, it now is clear that the manual inputs were fouled up and the model had other problems. But the temporary risk limit increase came off and everything looked normal as long as one did not know the facts.

Then, at the end of the first quarter of 2012, JP Morgan Chase filed a new 8K that reported the CIO’s VaR at $67 million, just $3 million more than the prior quarter’s figure. However, the bank did not disclose that the two numbers were not comparable. Investors had no way of knowing that the CIO was within risk limits solely because the bank changed the way that the figure was calculated.

It turns out that the CIO traders decided to double down as they were permitted to do since risk limits were no longer breached. Instead of winding down the obviously risky position, it ballooned from $51 billion in notional amount at the beginning of 2012 to $157 billion when the order finally came to stop trading at the end of March. As a result, the risk management controls now look even more inadequate than they would have had the bank just swallowed the bitter pill in January.

The story here is the incompetence of an organizational function that the bank had trumpeted as the best in the world. That is what the bank feared disclosing. And that is what they never could have reasonably considered to be a “tempest in a teapot.”