Back in late 2012 it came to light that Deutsche Bank may have hidden more than $10 billion in crisis-era paper losses on derivatives deals the bank struck with Canadian issuers of asset-backed commercial paper.
Essentially, the bank bought insurance policies on corporate credit from Canadian special purpose vehicles who were allowed to leverage the trades 10X. When credit spreads blew out in the wake of BNP Paribas’ move to freeze 3 of its ABS funds in August of 2007, the value of that protection ballooned, but because the bank had allowed its counterparties to leverage the trades, and because the non-recourse deals allowed the Canadian SPVs to simply walk away without meeting margin calls, the German lender was underwater to the tune of billions.
The loss, had it been recognized, would likely have forced the bank to seek a government bailout during the financial crisis, but Deutsche Bank essentially ignored it until several former employees blew the whistle (literally).
Now, Deutsche has reportedly settled the case for the paltry sum of $55 million and as with any other TBTF settlement, no actual people will be held accountable. Here’s WSJ with more:
Deutsche Bank AG has agreed to pay $55 million to the U.S. Securities and Exchange Commission to settle allegations it hid paper losses of more than $1.5 billion during the financial crisis that began in 2008.
The giant German lender said it didn’t admit or deny the allegations and that no charges were brought against individuals in the matter.
“The SEC acknowledged the bank’s cooperation throughout the investigation,” Deutsche Bank said in a statement Tuesday.
The SEC said in a separate statement Deutsche Bank has underestimated certain risks by between $1.5 billion and $3.3 billion…
The allegations behind the settlement announced Tuesday date to late 2008 and early 2009. A whistleblower at the time alleged the bank didn’t update the market value of certain credit default swap transactions, known as super senior trades. The whistleblower alleged the bank thereby masked mounting losses as the market value sank.
Deutsche Bank said Tuesday that it didn’t update the transactions’ market value because it believed at the time that there was no reliable method for measuring them amid illiquid market conditions during the crisis. The bank said it had since enhanced policies, procedures and internal controls regarding the valuation of illiquid assets.
Stefan Krause, an executive at Deutsche Bank who was finance chief at the time of the alleged mispricing, said in late 2013 that “the bank valuations and financial reporting were proper during the period in question and the allegations of fraud are wholly unfounded.”
In fact the allegations were not "wholly unfounded" at all. For those interested to know the whole story, including how the bank was able to escape with such a small fine, read on.
* * *
The ordeal dates back to 1998 when a former IBM leasing division Treasurer named Dean Tai linked up with two securities lawyers named Geoffrey Cornish and David Ellins to found Coventree Capital. Canada-based Coventree was a pioneer in the Canadian market for credit arbitrage-backed commercial paper. Here’s how the business model worked. Coventree -- and other Canadian “sponsors” -- set up special purpose vehicles (or “conduits” as they were called) which sold short-term commercial paper to investors in Canada via a network of dealers (usually banks). The conduits used the proceeds from this paper to finance bets in structured credit.
For example, some conduits made money by arbitraging the spread between what they could receive from selling protection on super senior credit and what they had to pay out on the notes they issued. In order to sell credit protection to banks however, the conduits needed to post collateral -- the collateral was funded by the sale of commercial paper. As long as the yield the conduits paid to commercial paper investors was less than the premium they collected from selling super senior credit protection to banks, the enterprise was profitable. Of course, super senior spreads weren’t very enticing back then, so the conduits needed a way to boost premium payments in order to make the whole thing worthwhile. Naturally, the investment banks on the other side of the trade (i.e. the banks buying the protection from the conduits) were ready with a solution. In so-called Leveraged Super Senior (LSS) deals, the conduits were only required to post collateral that was 10% of the notional amount insured. The more you insure, the higher the premium payments, so by allowing the conduits to lever up 10X, investment banks made sure the trades paid out enough to keep the conduits happy.
The conduits didn’t confine their exposure to synthetic super senior credit. Commercial paper investors’ money was used to finance all manner of “assets.” Take for instance Aria Trust, a conduit sponsored by a Coventree clone called Newshore Financial. Aria issued approximately $1.5 billion worth of notes backed by an eye-watering array of assets including (from the information statement issued by the Pan-Canadian Investors Committee) synthetic exposure to “two bespoke pools of investment grade corporate obligors, two CDX indices, one bespoke pool of investment grade corporate and sovereign obligors, and one sub-prime portfolio”, along with seven traditional programs comprised of “commercial mortgage-backed securities, notes referencing certain collateral debt securities, and auto leases.” The bottom line for the conduits was this: invest wherever the spread between what was being bought with commercial paper investors’ money and what needed to be paid out to those investors was high enough to earn a profit.
Note the glaring problem with this business model: Investors in the commercial paper needed to be paid within 30-90 days (this was short-term paper) but the assets their money was used to finance did not mature for years and so, if at any point the pool of new investors dried up, the whole enterprise collapsed as there simply was no way to repay all of the maturing commercial paper with the relatively illiquid assets owned by the conduits. In other words, the conduits were perpetually borrowing short to lend long. Enter “liquidity protection.”
In order for the conduits to receive an exemption from providing investors with a prospectus for the commercial paper (yes, you read that right, the commercial paper came with no prospectus despite the ridiculously complex nature of the assets backing the payment streams), the conduits had to procure an “approved” credit rating from an “approved” credit ratings firm. Canadian ratings agency DBRS (the only ratings agency that would touch this stuff) required issuing trusts to purchase liquidity protection to guard against the possibility that unforeseen circumstances could render the issuers unable to pay investors back. Again, this was necessary because the commercial paper came due every 30-90 days while the underlying assets (the auto loans, mortgages, structured finance programs, etc) didn't mature for years or, in some cases decades. While Moody's and S&P required what is known as “global-style liquidity” protection, DBRS required only that the issuing trusts purchase so-called “Canadian-style liquidity.”
Global-style liquidity arrangements required the liquidity provider to step-in and purchase maturing paper from investors if, for any reason, the trust which issued the debt was unable to pay when the commercial paper matured. The problem for the conduits in this scenario was that such ironclad protection came with a relatively high price tag. In Canada, trusts could save money by opting to purchase Canadian-style liquidity instead. This 'protection' was more cost-effective precisely because the liquidity providers knew that due to the way the Canadian-style agreements were worded, it was highly unlikely that they would ever be forced to make any payments. With Canadian-style liquidity protection, the liquidity provider did not have to make investors whole if it could be determined that a “general market disruption” had not occurred. The problem: it was the very banks providing the liquidity protection that got to determine the definition of “general market disruption.”
In fairness, this conception might be too strong. It seems the standard definition of a general market disruption revolved around the ability of the biggest players to sell their notes. If, for instance, issuers representing more than half of the total market could not place their paper, the market had suffered a “disruption event”. Of course this definition meant that a market disruption was exceedingly unlikely to occur. The so-called “Big 6” Canadian banks issued over half of all asset-backed commercial paper and these institutions were far less likely to have trouble rolling their paper in a crisis than a third-party issuer (i.e. the conduits). Thus the conduits’ ability to tap their liquidity lines in a crisis depended on whether the country’s largest financial institutions were able to place their notes. This was an apples-to-oranges comparison — investors would stop buying paper from the conduits well before they stopped buying bank-sponsored notes.
To illustrate the dynamics of this rather peculiar arrangement, imagine a situation wherein the cosigner on an auto loan gets to define the meaning of “default” by reference not to the person taking out the loan (i.e. the person whose payments the cosigner is guaranteeing) but rather to the entire universe of borrowers making car payments. If anyone, anywhere is current on their payments, the cosigner does not have to make the lender whole in the event the borrower defaults.
In sum, the market for Canadian third-party, asset backed commercial paper consisted of four types of participants: commercial paper issuers (the conduits), commercial paper distribution agents (sales teams at banks), liquidity providers (banks), and asset providers (banks). In one way or another, Deutsche Bank played each one of these roles.
For instance, Deutsche habitually signed Canadian-style liquidity agreements with the conduits, no doubt because the bank realized that the fees it earned from its role as liquidity provider (generally between 2 and 15 bps) were simply free money as long as the definition of the term “market disruption” was left to its discretion. In fact, the bank was the liquidity provider for nearly two thirds of all outstanding Series A notes issued by third-party sponsors (i.e. the conduits) in Canada. In other words, Deutsche Bank ostensibly guaranteed about $9.2 billion in commercial paper.
Similarly, Deutsche was the primary source for synthetic programs backing conduit-issued commercial paper. If you bought third-party commercial paper backed by an investment in a synthetic CDO in Canada prior to 2007, there was a very good chance that the synthetic program was structured by Deutsche Bank. More specifically, the bank provided assets for some 37 synthetic programs, more than double the number of programs furnished by the next closest investment bank.
To illustrate how large the bank loomed in the market, consider Apsley Trust (whose $900 million in residential mortgage-backed security-linked CDO transactions represented 49% of all conduits’ exposure to the cratering U.S. housing market) and Whitehall Trust (which issued $2.51 billion worth of paper backed exclusively by leveraged synthetic assets). Whitehall and Apsley, which issued their first asset-backed commercial paper on August 15, 2005 and November 24, 2005, respectively, were sponsored by Metcalfe & Mansfield, itself a subsidiary of Quanto Financial Group which was founded in October of 2005 by Mathieu Lafleur-Ayotte and Alain Pelchat, two managing directors at National Bank’s structured finance division. In the beginning, Lafleur-Ayotte and Pelchat each owned 20% of Quanto with the remaining 60% split between their former employer National Bank (15%), their other former employer Deutsche Bank (15%), and “key” employees (30%).
Given this, it comes as no surprise that National Bank and Deutsche Bank figured prominently in the subsequent issuance of notes by Metcalfe and Mansfield-sponsored conduits. In the information memorandums for Apsley Trust and Whitehall Trust, National Bank and Deutsche Bank are listed as distribution agents, whose job it was to “solicit and receive offers to purchase notes issued from time to time, arrange for the marketing and distribution of the notes, and … supply the Trust[s] with certain related advisory, investment, treasury management and administrative services.” It is certainly questionable whether National Bank and Deutsche Bank could have been reasonably expected to act in an unbiased manner when selling notes issued by a subsidiary of a company in which they held a combined 30% ownership stake and which was founded by two former employees of both firms.
With Deutsche Bank, the potential conflicts of interest ran even deeper than the bank’s role as a seller of the commercial paper. On top of the 15% stake the investment bank held in Quanto, Deutsche Bank served as the asset and liquidity provider for Apsley and Whitehall. Thus Deutsche owned part of the trust for which it provided assets and for which it served as a liquidity guarantor. Between this and the firm’s role as a distribution agent (through its securities division) Deutsche Bank: 1) provided the assets that backed the paper issued by Apsley and Whitehall, 2) signed liquidity agreements promising to provide payments to noteholders should there be a “general disruption” in the market for the paper backed by the assets it provided, 3) sold the paper Apsley and Whitehall issued, and 4) owned a 15% stake in Quanto, the parent company of Apsley and Whitehall’s sponsor, which was founded by two former Deutsche Bank employees.
To sum it up, Deutsche Bank was involved in every single aspect of this market. They had an equity stake in the parent of at least two issuers, they served as a liquidity provider on over half of all Series A commercial paper issued by Canadian conduits, they sold the paper through their securities division, and perhaps most importantly, they structured the programs (e.g. LSS deals) that backed the paper.
The events that unfolded between June of 2007 and October of 2007 are a story in and of themselves, but suffice to say that the market for commercial paper issued by the Canadian conduits imploded on August 13, 2007 (BNP’s move to freeze three ABS funds four days earlier sparked a panic) imperiling retail investors, small- to mid-size corporations, and pension funds and triggering a massive (and incredibly messy) restructuring effort.
As we’ve mentioned previously, Deutsche Bank played an outsized role in the market for LSS deals in the years leading up to the crisis. In fact, Deutsche Bank accounted for between $120 and $130 billion of the $200 billion (notional) in total LSS deals between 2005 and 2007.
As a refresher, here’s a simple explanation of the gap option problem with LSS deals:
The laughable thing about LSS deals was that they were effectively non-recourse, meaning that the protection seller was allowed to sell protection on a notional amount that was multiples of the collateral posted, but in the event the market moved against the seller enough to chew through that collateral and a margin call was made, that seller could just say “to hell with it” and walk away from the deal. More simply, I, the seller, insure $100 million in debt, but only post $10 million up front. If there’s a credit market meltdown and my $10 million is no longer sufficient and you, the protection (insurance) buyer, call me looking for more money to compensate you for the elevated risk, I can politely tell you to piss off. The risk that I tell you to piss off is called “gap risk.”
To be a bit more specific, the seller of protection (in this case the Canadian conduits) had the option to walk away from the deal without posting additional collateral (this is the “gap option”), and the value of that option changed depending on a number of factors including credit spreads and correlation.
As it turns out, Deutsche Bank began making these trades without even having a model to value the gap option —standard models (e.g. a copula model) cannot be used for LSS trades. Not only that, the bank’s credit correlation desk didn’t even bother to consult the market risk methodology department and instead decided to simply discount the value of the trades by 15%. Sensing that this was likely inadequate, Deutsche briefly attempted to determine the actual value of the gap option on the trades, but when the numbers came back looking rather nasty, the bank did what any pre-crisis sell side firm worth its salt would do: they scrapped that model and went with something that made the results look more favorable.In this case, Deutsche simply set up the equivalent of a loan loss reserve for the entire book and called it a day.
At the time (i.e. between 2007 and 2009), other players in the industry valued the gap option at between 2% and 8% of notional. Taking the midpoint there, and taking the midpoint between Deutsche’s estimated $110 and $120 billion in notional exposure, the value of the gap option for the bank would have been nearly $6 billion.
As for why Deutsche Bank was able to escape from the above by paying only $55 million to the SEC, consider that Robert Rice, then Deutsche’s Head of Governance, Litigation & Regulation for the Americas and William Johnson, Deutsche’s outside counsel both worked in the U.S. Attorney’s Office for the Southern District of New York with Mary Jo White and Robert Khuzami.
After his first stint in public service, Khuzami went on to become General Counsel to the Americas at Deutsche and by the time a former employee reported his concerns to the government in 2011, Khuzami had moved on to become Director of Enforcement for the SEC. Mary Jo White would of course become SEC Chair in 2013, and Rice would subsequently be named Chief Counsel to White.
In other words, a tight-knit faction of former attorneys for the Southern District of New York have managed to turn the SEC into an extension of Deutsche Bank, much as Goldman has turned the Fed into an extension of the Vampire Squid. As an aside, Deutsche’s General Counsel Richard Walker worked at the SEC for a decade and served as Director Of Enforcement from 1998 until his move to join the bank in 2001.