“THE GUY WITH A BOMB STRAPPED TO HIS CHEST”

“House prices are set to drop in the US for the first time on record, US investment bank Goldman Sachs warned this weekend.” With that statement from a July 29 article [1], we all entered a different world. No longer can Joe Sixpack count on constantly increasing home equity to bail him out of his financial problems. Any significant nationwide drop in house prices will inevitably force many homeowners into default. From now on, investment in mortgages is going to get riskier. And it’s not just individuals who will be affected. These changes will bring new challenges to the big GSEs that form much of the foundation of the US house finance industry. But a murky regulatory environment might well tie the hands of their risk managers at this sensitive time. Today’s post will explore this story.

Introduction

Anyone who saw this fascinating PBS report [2] last May is well aware of the ongoing accounting scandals at Fannie Mae. Much has been made of the famous “cookie jar” accounting used to move earnings from one year to another and trigger bonuses for top management. Regulators have estimated that these shenanigans, plus other errors mostly in accounting for financial derivatives, will result in a restatement of earnings that will show a loss of about $12 billion when the smoke clears. Most analysts feel this will be painful but can be absorbed by such a huge company. However, another aspect of the scandal has largely flown under the radar. Some journalists have suggested that this problem might actually pose an existential risk to the company. It’s Fannie’s QSPE trust funds, and that’s the thread we’ll be following.

The GSEs use derivatives to protect themselves against fluctuations in interest rates, but the risk of a wave of defaults is managed by a different tool. Of course the managers have always been prepared for such a possibility. The basic idea is to share the risk with other companies, much as reinsurance shares risk in the insurance industry. Unfortunately, the method the GSEs use for this sort of risk sharing was recently abused in a completely different industry. As a result, the accounting standards people have undertaken a long and arduous process of tightening and rewriting in this area. As the GSEs’ risk managers fine tune their defenses for the emerging challenge, they face an era of unparalleled murkiness in the relevant accounting guidelines. Welcome to the mystical world of Rule 140.

It’s All About Risk

Fannie Mae started up in 1938, during the Great Depression, because local US banks couldn’t handle the risk of supplying mortgages to their communities. Fannie, as part of the government, bought the banks’ existing mortgages and assumed their risk, so the local banks’ capital was freed up and they could originate more mortgages. Thus Fannie provided liquidity to the US home finance market at a time of great need. Fannie’s evolving charter has morphed the company away from being a creature of the US government but, like its little brother Freddie Mac, it has not yet been completely privatized.

Throughout its existence, Fannie has perfected the art of creating “baskets” of mortgages and then reselling them to investors as a large family of Mortgage Backed Security (MBS) bonds. This innovation made the company a pioneer in what is now a large securitization industry. The buyers of these MBS then assume most of the risk for the mortgages they represent, although Fannie guarantees the payments on these mortgages backed by its market capitalization of about $45.9 billon (Fannie is the second largest financial institution in the US behind CitiBank).

As Fannie has evolved into a more-or-less private company with real shareholders, profitability has become important. The company discovered it could make money by keeping a lot of the mortgages it buys. Fannie’s retained portfolio eventually grew huge by any standard, and currently stands at about $731 billion. Clearly, this lump of assets represents, at least potentially, a huge pile of concentrated risk. Had OFHEO, Fannie’s regulator, required 2.5% capital to back up all this stuff (instead of the 0.45% it requires for mortgages it merely guarantees [8]), it would have represented almost $15 billion. This could have hurt the shareholders, as Fannie’s profit might then have been divided out over a lot more stock.

Fannie’s managers work hard to maximize the benefits to the company and minimize the risks of their gigantic retained portfolio. Their preferred tool in this effort is the Qualifying Special Purpose Entity (QSPE) [3]. A Fannie QSPE is usually a basket of mortgages that the GSE has more-or-less “given away” to another financial institution (the counterparty, who might be a hedge fund or big bank). The counterparty assumes some of the ownership, and most of the control and risk for the basket. It’s sort of like a blind trust. Fannie removes most of the value of the QSPE basket from its own balance sheet and, Deo volente, the counterparty adds it to theirs. Sales accounting is the Holy Grail for a QSPE basket, because it allows Fannie to remove most of the value of the basket from their own balance sheet. This happens when the QSPE contract is deemed to be close enough to a real sale. On the other hand, the auditors and regulators may find after the fact that the basket doesn’t “qualify” after all, in which case the basket goes back (is consolidated) onto Fannie’s balance sheet. Thanks to the large amounts of money involved, the interpretation of these rules can be quite a sensitive issue, to say the least.

The Enron Factor

The occult world of QSPE management and sales accounting was proceeding quietly until the Enron scandal turned it upside down. Investigators found that senior management had been using QSPEs not for legitimate business purposes, but simply to hide lots of bad stuff from their investors, regulators, and auditors. The accounting community was none too pleased, and initiated a massive effort to make sure that such a thing would never, ever, happen again. Sadly, Fannie’s (and Freddie’s) risk managers were caught squarely in the crossfire. They depend critically on QSPE accounting to control the risk of the retained portfolio. This expert clearly shows how QSPEs have a valid role in this context.

“One difference [between Enron style finagling and legitimate securitization] is that securitization is normally used by an originator to obtain lower-cost financing through disintermediation, or removal of intermediaries such as bank lenders between the originator and the ultimate source of funds, the capital markets. This avoids the mark-up charged by a middleman of funds, and also enables the originator to raise funds more cheaply based on allocation of risks, assessed by parties having the most expertise. This is markedly different from Enron’s use of [QSPEs] for mere financial statement manipulation.

To the extent securitization is used to keep debt off an originator’s balance sheet, it superficially resembles Enron’s use of [QSPEs]. But there are important differences because securitization, unlike the Enron [QSPE] transactions, unambiguously transfers risk from the originator to the [QSPE] and its investors; and transfer of risk is, and should be, central to the accounting determination of non-consolidation.”

Schwarcz [4], p. 1551.

As can be seen, QSPEs are essential tools to get economies of scale and effectively manage risk for MBSs. In case it wasn’t obvious, “non-consolidation” was accountant-speak for “you don’t have to put the basket back on your balance sheet” (at least I think it was).

A Complicated Rule

QSPEs exist in an international environment. Some of the alphabet soup involved is captured in this 2003 powerpoint show [5] that explores things from an Australian perspective. The accounting for American companies like Fannie Mae is mostly controlled by the Financial Accounting Standards Board (FASB). QSPEs are governed by their Rule 140.[6] The point of this exercise is simply to tell when a transaction between two companies is a sale as opposed to a loan, but if you can make heads or tails of any of it you’re a better man than I am. As mentioned above, the fine points of the distinction can have enormous consequences for a company’s accounting.

The QSPE community is following with rapt attention FASB’s long running effort to clean up its post-Enron mess on Rule 140. Therefore, the following rather baffling story from the July 31, 2006 edition of The GSE Report is possibly more newsworthy than it appears.

“The Financial Accounting Standards Board tentatively decided to redeliberate three issues in its project on securitizations accounting, including rollovers of beneficial interests in late September and October. The board will readdress the issue of rollovers of beneficial interests and whether such practices should be permitted activities of particular special purpose entities that qualify for sale accounting. FASB will also weigh again the topic of participating interests and transferability requirements (paragraphs 8(a) and 9(b), respectively, of the pending draft amendment). (Bureau of National Affairs, 07/27/06)”

section titled: “FASB to redeliberate issues concerning securization accounting”

The BNA is a paid subscription, but a report of the FASB deliberations is available from the FASB site. 

“At the June 7, 2006 meeting, the Board continued redeliberations on the August 2005 Exposure Draft, Accounting for Transfers of Financial Assets, by discussing whether to proceed with this project or to reallocate resources to a long term project to replace Statement 140.”

FASB 22Jun06 [7]

 

These discussions are really complicated (even compared with Rule 133, the famous derivatives rule), but remember that one wrong move and these guys might blow away a good hunk of the capital of a major GSE (besides causing chaos among the many other industries that depend on QSPEs). Both Fannie and Freddie have yet to clean up their financial reporting. FASB’s ongoing discussions provide a moving target for the big GSEs as they get their financials to accord with Generally Accepted Accounting Principles (GAAP). Considering that the procedures embodied in 140 also guide Fannie’s risk management for its retained portfolio, it would be a good thing if FASB could get on with the job with all possible deliberate speed.

Close Call

In the course of their other investigations, Fannie’s regulators discovered to their shock that the company’s accountants had been surprisingly sloppy about documenting the company’s QSPEs. This appeared to be true even for the non-retained mortgages, the ones actually sold as MBSs. As unbelievable as it sounded, in the worst case the company would need up to $30 billion of additional capital [8], engulfing nearly the whole value of its outstanding stock. The very size of the problem made everyone take a deep breath and step carefully back from the brink.

“‘One Fannie [critic] says [Fannie Mae’s] best defense may be that the size of the issue means “Fannie is like the guy with a bomb strapped to his chest. The situation has to be treated delicately,”‘ writes Stephen Schurr in The Financial Times [London, England].”

The GSE Report, April 11, 2005.

The story became public in early April 2005, but was not very widely reported. Most of the analysis was devoted to speculation on how the problem would be resolved short of imploding the company, as Fannie is truly almost “too big to fail”. A typical scenario was this one.

“Of course, if Fannie’s QSPE accounting isn’t correct, it could in theory transfer the assets back on balance sheet and immediately reissue them with the correct accounting. But it wouldn’t be as easy as that. Investors would probably demand some fee for participation in an exchange offer. Moreover, it might be extremely difficult to do because the documentation may be too slipshod to do an organized reissue.

…

… if the value of the once-off-balance-sheet assets slid in an economic downturn and Fannie was holding less capital than it should against them, the company would be horribly exposed.”

Eavis [9] April 6, 2005

Things calmed down considerably after the spring of 2005. In what the pundits regard as a happy ending to the story, OFHEO’s recent report in May of this year estimated that only $25.5 billion of the QSPEs would need to go back on Fannie’s balance sheet.[10] That number requires just $600 million of additional capital, and is a final scenario that greatly exceeded analysts hopes.

We Are Not Quite Out Of The Woods Yet

To a certain extent, Fannie’s QSPE adventure so far has been a largely academic accounting event. More graphically, perhaps it’s like a military training exercise where the referee (the accountants and regulators) judged Fannie not to have “died” in the course of this story. They have passed a positive judgement on Fannie’s “readiness”, but the real war is about to begin. Fannie’s QSPEs provide a bulwark (like a system of levees) against a real surge of defaults. The nation wide house price declines predicted by Goldman Sachs [1] will most likely provide just such a surge. The QSPEs are designed to move the resulting losses around the financial system in such a way that the system as a whole survives. FASB, the GSEs, OFHEO, the banks and hedge funds have all expended a lot of thought in modeling the system for just such an eventuality. Within a year or so, we will all know if they prepared well.

________________________

Notes and References

[1]: “Goldman-Sachs Say Home Prices To Drop”, The Chattanoogan, posted July 29, 2006. The emphasis in all the quotes is mine.

[2]: “The Mortgage Giant Fannie Mae Accused of Deception and Mismanagement”, OFHEO Head James Lockhart interviewed by Jeffrey Brown, Jim Lehrer Newshour, May 23, 2006.

[3]: QSPEs are often called SPEs when qualification, the Q, is under discussion. Fannie’s QSPEs are often termed QSPE Trust Funds. An equivalent formulation is Qualifying Special Purpose Vehicle (QSPV), SPV, or just “vehicle”. I’ll be using QSPE in this discussion.

[4]: “Securitization Post-Enron”, by Steven L. Schwarcz, Cardozo Law Review, April 2004. pp 1539-1575. Schwarcz uses the equivalent “SPV” instead of QSPE in the actual article.

[5]: “Do You Need To Consolidate (PPT)”, by Stephen Gustafson, Partner, Deloitte Touche Tohmatsu, Australian Securitization Forum, 30 April 2003.

[6]: Rule 140 is often called Financial Accounting Standard (FAS) 140, Statement of Financial Accounting Standards (SFAS) 140, FASB 140, or even SFAS 140/156 (because Rule 156 is a work-in-progress ammendment to 140).

[7]: “Project Updates: Transfers of Financial Assets (Proposed Amendment of FASB Statement No. 140, Accounting for Transfers and Servicing of Financial Assets and Extinguishments of Liabilities)”, FASB, Last Updated: August 3, 2006.

[8]: “Trust problem could force Fannie to find billions”, Reuters, April 6, 2005 (reposted by LibertyPost)

[9]: “Fannie Mae’s Trust Fund Troubles”, by Peter Eavis, TheStreet, April 6, 2005.

[10]: “Fannie Mae finds more accounting errors”, By James Tyson Bloomberg News, May 10, 2006.