Give me
Your dirty love
Like you might surrender
To some dragon in your dreams
Give me
Your dirty love
Like a pink donation
To the dragon in your dreams
I don't need your sweet devotion
I don't want your cheap emotion
Just whip me up some dragon lotion
For your dirty love
“Dirty Love”
Overnight Sensation (1973)
Frank Zappa
Updated -- A number of commentators argue that the tendency of government to bail out large banks, the institutions we lovingly refer to as the “zombie dance queens,” remains intact, this despite legislation such as the 2010 Dodd-Frank law. But such views may not adequately take into consideration why the zombie girls got so big in the first place. This comment is based on a paper on the “Shadow Banking System” to be published later this year by Indiana State University. -- Chris
Simply stated, the largest commercial banks became “too big to fail” in large part because they used non-bank vehicles to increase leverage without disclosure or capital backing. Their intent was to reduce the apparent capital needs of banks. Banks’ abuse of non-bank vehicles to issue subprime securities and hide capital deficits was facilitated by legal counsel, auditors, rating agencies and regulators, who all pretended that four centuries of legal precedent regarding financial fraud had somehow never occurred. Until 2011, FDIC rules did not preclude that abuse and even sheltered banks from need to disclose it to auditors and investors.
The failure of Lehman Brothers, Bear Stearns and most notably Citigroup all were largely attributable to deliberate acts of securities fraud whereby assets were “sold” to investors via non-bank financial vehicles. These transactions were styled as “sales” in an effort to meet applicable accounting rules, but were in fact bank frauds that must, by GAAP and law applicable to non-banks since 1997, be reported as secured borrowings. Under legal tests stretching from 16th Century UK law to the Uniform Fraudulent Transfer Act of the 1980s, virtually none of the mortgage backed securities deals of the 2000s met the test of a true sale. Under the UFTA standard, for example, any transfer which is intended to leave the transferor with insufficient capital is a fraud which converts a “sale” into a “secured borrowing” by the transferor.
Since the purpose of most bank asset "sales" via securitization was always "capital relief," no honest lawyer could say that the transfers met the UFTA standard applied to non-banks. Banks avoided balance sheet treatment for securitizations merely by "purporting to sell" loans to trusts. Bank regulators allowed theses “off-balance sheet” vehicles to be excluded for the purposes of determining regulatory capital requirements. When the crisis hit, it suddenly became clear that the banks’ capital was insufficient.
Today much of the “shadow banking” system with respect to residential real estate has run off or is in the process of doing so, but hundreds of billions in claims against banks arising from these purported “sales” of assets remain pending before the courts. Perhaps more important than prosecuting past acts of fraud connected to the creation and sale of bad securities, we need to clarify, going forward, what it really means to create a “sale” in the context of transfers made in an ABS transaction. Unless and until we do that, the ability of banks to generate “off balance sheet liabilities” that de-stabilize financial markets and cause crises will just morph to new forms. TBTF will remain alive and well for the zombie banks.
As noted earlier, bank abuses of non-bank vehicles to pretend to sell assets and thereby lower required capital levels was a major cause of the subprime financial crisis. It now appears that a major catalyst for this came in the wake of accounting changes that took effect in 1997. That’s when the FDIC adopted a “safe harbor” for bank securitizations to assure certain asset sales would comply with the new accounting principles.
Thus, most of the securitizations done by banks over the past two decades were in fact secured borrowings, not true sales, and thus potential frauds on insured depositories. In 2008, the Institutional Risk Analyst ran an interview with Professor Joseph Mason of Louisiana State University about the “good sale” issue. He noted:
“The subprime crisis results from a growing arbitrage of regulations and accounting rules that got out of hand. Right now the situation is that regulators don't want to acknowledge the problems in the market because to do so is to admit that they missed these same problems, in some cases going back 30 plus years now… When I joined the OCC in 1995, I focused on securitization as an area of research. Securitization was growing by leaps and bounds and clearly had safety and soundness implications. Nobody was collecting information on it. The general view of securitization was "loans are sold" and this process was viewed as a good thing. But I began to look at the fact that securitization was a funding mechanism for banks. It is the liability side, the funding side of banks, where you really are running into risk. Incidentally, the liability side is where mark-to-market accounting and fair value is getting into trouble, posting paper gains when liability values plummet in distress. We allow banks to fund themselves by selling securitizations into an illiquid market. If that illiquid market breaks down, then your entire bank intermediation system gets a hiccup.”
Mason’s comments in 2008 were entirely on target regarding the threat from securitization and how the zombie banks used off-balance sheet finance to grow their leverage and risk to several times their actual balance sheets. Mason’s comments also document the indifference of regulators like the Office of the Comptroller of the Currency to the problems created by badly constructed bank securitizations:
“Back in 1995 I was looking at this rapid growth in securitization and I began to suggest to my colleagues at OCC that we should start to monitor the phenomenon. I got tremendous indifference. This was around the time of the Advanta failure. I knew that there were some things going on within the agency because this failed bank did not have any deposits and there was some uncertainty whether the FDIC would liquidate the bank or just leave it to the OCC. But more than the question of the receiver, the Advanta portfolio was so securitized and so heavily levered that nobody knew how to value the business. When the bank was sold to Fleet, it was transferred as a whole business securitization where the buyer simply bought a majority share of the trust.”
Mason’s comments about the Advanta transaction illustrate the serious issues created when a bank sponsor creates hidden liabilities via a securitization:
“As I researched Advanta and other early examples of securitization it became clear that while we publicly toed the line about the validity of "true sales" when it came to securitizations, the reality was that these were anything but. Deals would sometimes run into legal problems - things like deals not accumulating sufficient reserves in the early stages to provide a buffer - which would be grounds for the deal to be called off entirely. While an intervention is a violation of the regulatory interpretation of true sale, nearly every time a securitization deal got into trouble the regulators would allow the banks to make collateral changes to these deals on the fly to make them work. This type of implicit recourse, especially in the world of credit cards and other types of consumer collateral, was the basis for much of my early academic research in the field.”
A year before our interview, Professor Mason and my friend Josh Rosner authored a prescient paper which illustrated the legal uncertainties in the legal definition of a “true sale” entitled “Where Did the Risk Go? How Misapplied Bond Ratings Cause Mortgage Backed Securities and Collateralized Debt Obligation Market Disruptions.” [26]Those of you who are big fans of S&P, Moodys and the other ratings agencies involved in fomenting the sub-prime crisis will especially appreciate this excerpt:
“In December 2000, LTV Steel filed for voluntary Bankruptcy protection under Chapter 11 in the US Bankruptcy Court of Northern Ohio120. In their filing the Company asked the court to grant an emergency motion to allow them to use the collections from the securitizations and claimed that the transactions were not “true sales” but rather “disguised financings”. The Court granted the Company’s motion though it did not rule whether or not the securitizations were “true sales”. Although this case could have caused the rating agencies to take the same position as the Georgia law, of ambiguity making it difficult to rate the risks to noteholders they chose not to. In fact, one of the agencies appeared to pressure attorneys to avoid commenting on the matter in legal opinions. Standard & Poor's insisted that attorneys submitting true-sale opinions to the rating agency stop referring to LTV, noting that the court never made a final decision and that such citations inappropriately cast doubt on the opinion. Seven months later, in a delicately worded press release, S&P withdrew that prohibition—apparently because lawyers refused to ignore such an obvious legal land mine.”
Under GAAP rules, after 1996 any transfer of a financial asset that could be unwound by a bankruptcy trustee or a receiver was required to be accounted for “on balance sheet,” as a secured borrowing. That had a particularly large impact on banks. By its governing law, FDIC as a receiver has arguable power to unwind any contract, and FDIC had never enacted a rule limiting that right.
So, FDIC found itself faced with a choice of precluding all financial asset sales by banks or defining limits on its rights in receivership. FDIC debated and adopted a “safe harbor” that gave assured sale treatment for certain transactions. With post-crisis hindsight, many such transactions by banks would, without that safe harbor, be deemed fraudulent. And now you know why non-banks such as Lehman Brothers used FDIC insured depositories as the conduits for selling residential mortgage backed securities (RMBS) to investors.
Until FDIC’s safe harbor rule was changed in September 2011, therefore, it can be argued that the FDIC waived a right to challenge certain fraudulent transfers by banks. During that period, the rule gave bank lawyers a safe harbor by which to provide opinions that securitizations were “true sales” even though we now know that was fiction.
The process of changing GAAP began when an earlier accumulation of “off balance sheet liabilities” by banks contributed to the real estate crisis in 1989-1992. That led the Financial Accounting Standards Board to adopt SFAS 125 which created a mechanism requiring lawyers to provide “true sale” opinions in the context of a securitization. The primary example FASB looked to was a requirement for opinions on "fraudulent transfer" law that major rating agencies demanded before they would rate any securitization higher than the rating of its sponsor/originator.
The reason FASB adopted the standard is simple: Failure to comply with fraudulent transfer law converts a financial asset “sale” into a “secured borrowing” to the extent that the transferee, in good faith, compensates the transferor. Rating firms such as S&P wanted comfort from the lawyers that the assets in an RMBS, for example, were beyond the reach of the bankruptcy court in the case of non-banks and FDIC, as receiver, in the case of a failed bank.
With this as FASB’s standard, auditors gained a benchmark that is not subject to manipulation. The line is set by law. Failure to comply with fraudulent transfer law creates a secured loan and compliance creates a sale. By the FDIC safe harbor adopted in 1997, however, banks were not required to meet the same test as non-banks, creating an imbalance of competition favoring the largest banks. The zombie dance queens exploited this legal loophole and thereby grew their risk profiles many times over. TBTF became inevitable.
After extensive review, in September 2011 (effective November 2011), the FDIC abandoned its former safe harbor and enacted a new rule that corresponds with the treatment FASB accords non-banks. That change finally establishes the “level playing field” which economists and experts have sought for US financial markets since at least 1969. To be “sales” of assets, all future securitizations will require analysis of whether a fraudulent transfer has occurred. Few observers have taken note of this development. Auditors will, it appears, follow suit and adopt new guidance which will effectively end the ability of banks to assert that a fraudulent transfer of assets can be accounted for as a “sale,” as has been the case for all non-banks since 1997.
The upshot of the rule change by the FDIC in 2011 is that banks will either have to change practices or they may not be able to meet the test for a “true sale.” If they fail the test, banks may instead need to keep mortgage or other exposures on balance sheet, supported by whatever capital regulators establish for those transactions. Good luck getting lawyers to opine in writing regarding a “true sale” transaction by zombie bank. This development could mean less finance available from banks for the mortgage industry, but it may also herald the end of “too big to fail” with respect to the largest banks. More important, it may lead to renewed growth of non-bank participation in a “true” and more stable shadow banking system. And that will be a good thing.
Endnote: In the sevice of the Bair kitchen cabinet, i conveyed a message to the supporters of covered bond legislation that 102% "OC" or overcollateralization was all that the agency could tolerate for these secured borrowings. If the House adopted legislation sponsored by NJ republican Scott Garrett, the FDIC official told me, the agency would oppose the bill and kill it in the Senate. Now I understand why. Anything above 2% capital buffer, absent the safe harbor would have violated the pre-1997 "true sale" rule. And we must note that FDIC, of all the regulatory agencies, has been the most attentive to the true sale issue and, now, finally got it right.
www.rcwhalen.com [27]