As Congress and the Federal Deposit Insurance Corp. (FDIC) debate the severity of restrictions to place on greedy lenders and Wall Street "securitizers" to ensure that they can never again jeopardize the global financial system while forcing droves of decent American families into homeless shelters, we would all do well to take a step back and consider where the U.S. housing finance system has been and where it is headed.
We may well conclude that, like a wilderness plane crash survivor trekking for days without a compass, we are likely to end up right where we started.
Until the entry of Fannie Mae, Freddie Mac and Ginnie Mae into the mortgage markets in the early 1970s and Wall Street in the late 1970s, American bankers accepted the savings of their customers and loaned the money to other customers as mortgage, commercial or consumer loans at an interest rate higher than the rate they paid their depositors and, after paying expenses, pocketed the difference. This is the very essence of banking.
Unless a bank's deposits grew substantially and provided new funds to lend, the banker often would wait anywhere from five to 30 years for a borrower to refinance, move or repay the loan at maturity, and then lend the repaid amounts to another customer.
Sometimes bankers would sell loans to, and buy loans from, their peers in other parts of the country to achieve greater geographic diversification, but once sold the loans usually still stayed on the other bank's balance sheet until the borrower repaid the loan.
By the 1970s, Americans were fed up with the slowness of this process and complained that banks' tight lending policies kept needed financing from deserving American families and prevented them from achieving the American Dream. Heeding this now once again familiar popular cry, Congress established the GSEs to increase the speed at which lent funds would recycle into funds for new mortgage loans.
The GSE model proved successful in increasing the funds available for loans under their "loan limits," and Wall Street investment bankers quickly copied the model for larger loans by securitizing them and selling them to institutional investors around the world. Banks increasingly moved from holding loans in portfolio to selling them upon origination to the GSEs or Wall Street, taking a small profit, and relending the funds to originate another loan for sale, and so on, and so on.
This "originate-to-sell" business model, and the increasingly sophisticated GSE and Wall Street securitization techniques it engendered, gradually overtook the "portfolio" business model and channeled vast amounts of private capital from around the world to the U.S. mortgage and consumer credit markets. This had the laudable side effect, from a bank's standpoint, of spreading the lender's credit risk onto the backs of non-bank institutional investors worldwide. The originate-to-sell model was the engine of the U. S. mortgage and consumer credit markets for two decades, and undoubtedly contributed significantly to the affluence of the U.S. homeowner and consumer.
In the mid-2000s, however, the accelerator was pushed to the floor and remained stuck, likely as a result of a Federal Reserve low-interest rate policy designed to remedy the effects of the dot-com crash, and funds were recycled through this capital markets machine with increasing and ultimately uncontrollable velocity, fueling rapid price appreciation in the U.S. housing stock and other assets.
When the bubble induced by the racing global securitization machine and easy money burst in 2007 and 2008, politicians and regulators were quick to blame the "originate to sell" model which, they said, did not incentivize lenders to make high quality loans because they did not suffer the consequences of poor loan underwriting decisions. The conventional wisdom is that lenders did not have sufficient "skin in the game," and that banks and other lenders were knowingly foisting low quality loans onto unsuspecting investors.
Congress and the FDIC now propose to remedy this failing with, among many other restrictions, the requirement that securitizers retain at least a 5% interest in loans they securitize.
Never mind that the securitization machine was fueled as much by insatiable investor demand for high yields and more product as it was by any pushing of product from originators. Never mind that the securities were highly rated by Standard & Poor's, Moody's Investors Service and the other international credit rating agencies, which were at that time the venerable hallmarks of impartial and rigorous credit analysis. Never mind the enviable performance of these instruments over a period of considerable years. Never mind that investors, bond insurers, accountants or rating agencies did loan-level due diligence on most securitization pools. Never mind that investors and rating agencies knew full well the risks of "stated income loans," "teaser rates," "interest only loans," "pay-option ARMs" and so on, and that the presence of these ultimately fatal defects in a loan pool was not hidden in loan files but was fully disclosed with statistical breakdowns in prospectuses complying with strict Securities and Exchange Commission rules. Never mind that most securitization investors were sophisticated institutional investors adept at the complex analysis required to value these products. And never mind that lenders made extensive representations and warranties regarding their underwriting practices, and were required to repurchase any loan that went bad because of a breach.
Even if one explains away most or all of these "never minds," there is one central, but rarely mentioned, fact that calls into serious question the validity of the "skin in the game" hypothesis. Specifically, many of the large bank and nonbank securitizers in the recent cycle, including Countrywide, IndyMac, Washington Mutual and New Century Financial Corp., often retained the most subordinated interests in their securitizations because they could not be readily sold at a profit. As a result, many securitizers already held first loss positions in the assets they originated, which is even more "skin in the game" than the pari passu interests contemplated by Congressional proposals and the FDIC. Indeed, write-downs and losses on these interests have contributed to multiple failures both within and outside the banking system.
One reason that these retained interests are rarely mentioned in the debate may be that they appeared on the balance sheets of the securitizers in relatively small amounts, rarely in excess of one percent of total assets. What is overlooked, however, is that, in accordance with the generally accepted accounting principles or GAAP, these interests were recorded at their "fair values," a measure that presumably took into account their virtual worthlessness as a result of their first loss position. If viewed in relation to the related pool's principal balance, it is quite possible that the retained percentages, in terms of principal balance, were equal to greater than the 5% retention benchmark under discussion by the rulemakers.
The bursting of the U.S. housing bubble and consequent national and global economic slowdowns has resulted in the demonization of securitization and a rash of initiatives by politicians, regulators and even accountants to prevent a recurrence of the meltdown, more often than not focusing on some aspect of the "originate to sell" model or the "skin in the game" concept. Many of the initiatives have a decidedly punitive feel. On December 11, 2009, the U.S. House of Representatives passed the Wall Street Reform and Consumer Protection Act of 2009. Among other reforms, the House bill would generally require sellers and securitizers of loans to retain a 5% interest in the loans, the risk of which may not be hedged. A companion Senate bill, unveiled by Senator Christopher Dodd, D-Conn.,in March 2010, also calls for a 5% retention.
Not to be outdone, on December 15, 2009, the FDIC Board issued an "advance notice of proposed rulemaking" on securitization reform, containing a number of "sample" restrictions and requirements that the FDIC is considering imposing on all U.S. bank lenders and securitizers.
The sample provisions, which contain the same unhedged 5% "skin in the game" risk retention included in the House and Senate bills, go considerably beyond the legislative proposals. Among other things, they include a requirement that securitized mortgage loans be "seasoned," or in existence, for at least 12 months prior to securitization, essentially precluding the use of the "originate to sell" model by U.S. financial institutions. They also prohibit the use of a financial guaranty or bond insurance as a means of credit enhancing securitizations for the benefit of investors, a device which historically lowered the cost of securitization and spread credit risk outside of the banking system.
The FDIC is charged with conserving the FDIC insurance fund and, therefore, assuring the safety and soundness of U.S. banks. Perversely, most of the "skin in the game" proposals proposed by the FDIC have the effect pushing credit risk, which prior to the meltdown was widely distributed among capital markets participants, back onto the banking system.
Of probably far greater ultimate impact than the Congressional and FDIC actions, in June 2009, the Financial Accounting Standards Board adopted new accounting rules, FAS 166 and FAS 167, which became effective on Jan. 1 for most securitizers, and required that many securitization structures previously treated as sales be accounted for instead as secured borrowings and consolidated onto the balance sheets of the securitizer, with the result that 100% of the assets in the affected structures will remain on-balance sheet. Ironically, it is likely that these new accounting rules alone, even without any help from Congress or the FDIC, will be enough to substantially shut down the U.S. securitization markets.
Although they allow for the possibility of off-balance sheet deals that would permit lenders to book an immediate profit and recycle funds for new lending, the new rules make this outcome so cumbersome and expensive that most lenders are likely simply to retain their loans in portfolio. This, in turn, will continue to severely limit the availability of funds for new residential lending.
This limitation results from the simple fact that assets retained on-balance sheet must be supported by capital. Banks and other lenders are already capital-challenged in the current environment, and many of the pending financial reform proposals would impose even higher capital requirements. Under current regulatory capital and leverage requirements, banks will be unable to significantly increase their retained financial assets without substantial increases in their capital, which are usually difficult and costly to achieve.
To be sure, banks may be able to raise some funds for lending by borrowing against their portfolio loans by pledging them to a Federal Home Loan Bank, issuing "covered bonds" or even undertaking on-balance sheet securitizations. However, in all of these cases the loans remain on-balance sheet and their growth is substantially constrained by capital requirements.
The cumulative effect of the pending legislation, the new accounting rules and the FDIC's rulemaking, even if ultimately softened, will be to return U.S. banks to the portfolio lending model that prevailed in the 1970s. Not only will this result in the desired retention by banks of a greater portion of the credit risk of financial assets, but it eliminates the considerable benefits of recycling loan origination funds that benefitted the U.S. economy for over three decades. When one considers that there are serious calls to substantially downsize or even eliminate the GSEs, it is evident that we are truly returning to our starting point - a point which even then was considered untenable because of the banking system's inability to provide sufficient credit to support a healthy residential mortgage market.
The policymakers' insistence upon adopting measures that will undo the risk-spreading and capital-attracting virtues of securitization is oddly reminiscent of Cleavon Little pointing a pistol at his own head in Mel Brooks' Blazing Saddles (1974).