Back to Portfolio for the Future™

A Close Look at the Deutsche Bank Findings

DBIn the course of a busy news cycle, this tidbit was largely overlooked: in June 2014, Deutsche Bank raised $11.6 billion in capital in part by offering shareholders a bargain-basement deal on its equity, a 24% discount on purchase of additional shares.

Such lengths indicated how badly it needed to recapitalize. Why?

A year after that recapitalization, the Securities and Exchange Commission shed some light on this question when it charged Deutsche Bank AG with misstating financial data, failing to take into account the risk of loss inherent in its leveraged super senior trades. The unravelling of some make-believe book keeping is what had in due course required the recapitalization.

In accord with customary (though much-criticized) practice, the SEC allowed the bank to settle the action “without admitting or denying the findings” of fact, except for the purpose of agreeing to the cease-and-desist order and civil penalties.

Let’s indulge ourselves in the temptation to do some rubber necking at the scene of this pile-up.

Buying A Safe Harbor

DB bought credit protection from Canadian counterparties (CC). These were bespoke synthetic collateralized debt obligations. The CCs posted collateral of about 9% of the total value of the trades, which is to say, the trades were leveraged at x11.

The trades increased in value as the markets deteriorated in 2008. After all, DB had bought itself a safe harbor as the ocean was getting stormy, and the storms made the safe harbor ever more precious. But … there was that x11 leverage, or the “gap risk,” which should logically have been a key component in the valuation of the LSS trades.

During the period [late 2007-08] DB used what the SEC calls “a variety of methods to attempt to measure the Gap Risk” in the LSS trades. The constant in those methods? – each change increased the value of DB’s position by decreasing the significance of that risk.

The Road to $0

The first method? DB measured the risk as a 15% haircut from the value of the LSS positions. As of December 31, 2007, this meant in absolute terms a haircut of $200 million. BUT … DB stopped updating that number thereafter. The value of the positions continued to increase, while the haircut stayed at $200 million, so the percentage value of that haircut of course declined.

In March 2008, the DB Market Risk Management group announced a new model for valuation on this point. But this, what the SEC’s Findings call the MEM model, was never implemented. It was supplanted immediately by the Desk Model, which set the Gap Risk at $20 million, or just 1/10th of what it had been in absolute terms at the end of the previous year. Still, one good feature of the Desk Model (so called because it was a brainchild of the trading desk) was that it was not static.

The Desk Model dictated a haircut that turned out to increase in the volatile weeks that followed. By the time of the Lehman bankruptcy, this Desk Model was requiring a haircut of $78 million.

In October 2008, the Gap Risk was in essence set at $0: hitting the “off” switch on the Desk Model and placing it with … nothing.

From Both Sides Now

The SEC found that IAS 39 was the relevant accounting standard for the LSS positions, which defines fair value as “the amount for which an asset could be exchanged, or a liability settled, between knowledgeable, willing parties in an arm’s length transaction.” This requires the assessment of all relevant available market information.

So the key language in the Findings is as follows: “Deutsche Bank recognized in its financial statements the increasing value of the Positions driven principally by increasing credit spreads and their increased volatility but, at the same time, did not take into account the impact of this same market information when measuring the Gap Risk in its LSS Positions.”

Judy Collins might suggest looking at risk from ‘both sides now.’ But according to DB at a critical moment in global financial history, risk existed only to the extent that it worked to enhance the value of DB positions: it didn’t exist in any sense that might have required a haircut.

Largely by virtue of this insistence on looking at clouds from only one side, DB violated section 13(a) of the Securities Exchange Act and rules 13a-1, 13a-16, and 12b-20 enacted by the SEC thereunder.

The “cease and desist” order means, in biblical terms, “go and sin no more.” [KJV John 8:11]. This is accompanied by a payment of $55 million for the unadmitted-to violations that DB promises to cease, payable by certified check or bank cashier’s check.