Financial Engineering and the Banks

Speculators may do no harm as bubbles on a steady stream of enterprise. But the position is serious when enterprise becomes the bubble on a whirlpool of speculation. When the capital development of a country becomes a by-product of the activities of a casino, the job is likely to be ill-done.

J.M. Keynes

The General Theory (1935)

Chapter 12

Assume that we have demonstrated that destabilizing speculation can occur in a world of individuals whom it is convenient and fruitful to consider as normally rational. Permit this world to be perturbed by a “displacement” of one sort or another, largely from outside the system, giving rise to prospects that individuals midjudge either for themselves or for others. At some stage, investment for use gives way to buying and selling for profit. How likely is the speculation to lead to trouble?

Charles P. Kindleberger

Manias, Panics, and Crashes

(2000 ed.)

An article in the 10/12/09 WSJ described the travails of the Fontainebleau casino in Las Vegas. After the sponsors and lenders spent $2 billion, the 4,000 room project will require at least another $1 billion to complete. The trouble is the casino, if completed, will be worth only $1.76 billion.

As the financial system deals with the consequences of failed commercial real estate projects and empty houses baking in the desert, we ask how this massive misallocation of resources was possible. Excessive abstraction, enabling large-scale markets, superimposed upon some very bad lending, that gave credit to marginal borrowers under the assumption that house prices (the collateral) would appreciate forever, was at the heart of the Great Financial Crisis of 2008. In the following, we discuss the abstract paradigm of modern loan portfolio risk management, the principles of sound lending, and how the financial system can allocate capital to its best uses.

Modern Loan Portfolio Risk Management

A major paradigm of modern financial risk management states that life is equivalent to a gamble in a casino, where all outcomes can be pre-specified and then quantified. We look into the black box of financial engineering to analyze the statistical models that were used to construct the structured financings that caused the current crisis. We discuss the strengths and weaknesses of the binomial and Gaussian copula models, and ask how their theoretical shortcomings practically mattered. As always, if your utility function does not include quantitative analysis, we suggest skipping to the paragraph marked *.

The Binomial Expansion Method

In 1996, Moody’s introduced the binomial expansion (BET) method for evaluating collateralized loan and bond obligations, CLOs and CBOs. This method priced an actual loan portfolio against an (analytically tractable) fictitious reference portfolio of bonds, whose defaults are assumed statistically independent. The assumed number of D bonds is a function of (the number of issuers in the collateral pool, industry sectors, and the actual loan amount relative to the average). Each theoretical bond is then assumed to have the same value and probability of default, determined by the weighted analysis of individual credits within the original pool.

The expected losses of this fictitious reference pool are then calculated according to the binomial formula that calculates the probability of scenarios ranging from 1 to D defaults. If ( j) is the number of defaults then:

P_j = D!/j!(D-j)! x p^j(1-p)^D-j

where: P_j = The probability of the scenario, j defaults.

D = the number of bonds in the theoretical portfolio.

! = factorial, so if there are three bonds D! = 1 x 2 x 3

p = the probability, equal for all theoretical bonds, of a single default

Having calculated the probability of (j) loan defaults, it is possible to assign an expected loss and return to each tranche (slice) of a structured financing, assuming some recovery rate. The binomial method for calculating expected tranche losses is rather straightforward. It suffers, however, from two drawbacks:

1) Like most quantitative analyses, this method requires the input of historic data to determine the default rates of the individual credits within the original pool. The results of the analysis depend upon the historic period chosen.

2) The statistical independence assumption assumes that one default is independent of the other, in spite of the common dependence of all defaults upon macroeconomic conditions.

How did this (relatively) simple binomial model work out in practice? Actually, not so bad. Moody’s applied this formula to structured finance analysis in 1996; and this model, and its variants, were in use until 2004 when Moody’s and S&P started using the Gaussian copula. The reason there were no major problems in structured finance manifest before 2007, despite the drawbacks above, was a totally necessary condition: the loan standards of credits written much earlier were higher.

The Gaussian Copula

The Gaussian copula was intended as an improvement over the BET method, allowing an explicit inclusion of asset correlation when calculating their default probabilities P_j , as in the above. The formula is complicated, but the idea is simple. Take two bonds, each with the independent probabilities of default of 5%. The joint probability of both defaulting is therefore .05 x .05 = .25%. However, if both bonds have a default correlation of .3, the joint probability of both defaulting is increased to .71%, as calculated by the Gaussian copula. Note that the joint probability of default is much higher than under the assumption of statistical independence. In 2005, (Garcia, Dwyspelaere, Leonard, et al.) compared the BET and Gaussian Copula models using Moody’s loan default data. They concluded, “A consequence of this change in (model) sensitivity due to correlation is that it becomes quite an art to interpret the results given by the (copula) model.¹ In general one might say that BET will give losses that are lower (our emphasis) than the copulas with the high correlation assumption…”

The copula correlation method could result in more conservative structured credit calculations. So what went wrong? We could note these theoretical problems with the copula model:

1) Financial data does not have a normal Gaussian distribution.

2) Asset correlations are not a constant, and approach one during a financial panic.

But the main problem with the copula model was its use. In 1999 David Li published a paper called, “On Default Correlation: A Copula Function Approach.” ²

Assuming that markets correctly price financial assets (at least for the moment), Li used credit default swap spreads, rather than actual loan default data, to determine the probability of any (n) members in a credit portfolio defaulting. To quote a March, 2009 Wired magazine article, “The effect (of this) on the securitization market was electric. Armed with Li’s formula, Wall Street quants saw a new world of possibilities. And the first thing they did was to start creating a huge number of brand-new triple-A securities. Using Li’s copula approach meant that ratings agencies like Moody’s – or anybody wanting to model the risk of a tranche – no longer needed to puzzle over the underlying securities. All they needed was that correlation number (our note: γ in the below, measured from not too stable asset correlation data), and out would come a rating telling them how safe or risky the tranche was. As a result, just about anything could be bundled and turned into a triple-A bond - corporate bonds, bank loans, mortgage-backed securities, whatever you liked…” To use the Wired magazine example:

-------The Gaussian Copula-------

Joint Probability of Two Defaults

over One Year

= Pr[T_A <1, T_B <1] = ϕ₂ (ϕ^-1(F_A(1)),ϕ^-1(F_B(1)), γ)

where: ϕ₂ = the bivariate normal distribution to be calculated, ϕ is the univariate

F_A,B = probability distribution of the survival times of asset A and asset B to

credit default, calculated from the credit default swap spread

γ = gamma, the correlation number derived from asset correlation data.

The estimation of γ is very complicated. Analysts use either ratings-

based or market-based measures. Those looking for Newtonian

predictability from the copula will be disappointed because,

according to Derivatives Week, “…there has not been consensus

among market participants on which method to use…” ³

In spite of all its Greek letters, financial engineering is actually

an art rather than a science.

This formula enabled anything with a credit default swap (insurance) spread to be placed, modeled, and then tranched in a generic collateralized debt obligation (CDO) financing.

* The Real World

The above indicates that structured finance, per se, was not solely responsible for the Credit Crisis of 2008. Recall that despite their theoretical drawbacks, BET financings did not blow up on a large scale. The Gaussian copula, moreover, in general led to higher loss estimates than the preceding. What was nearly fatal to the world’s financial system was the perfect storm of really bad mortgage and corporate lending ⁴, structured financings priced according to a credit market based Gaussian copula, sloppy implementation of this questionable model at the ratings agencies ^4a, and the tangled admixture of tranches combined with other tranches that was used to create more CDOs out of other CDOs ⁵. As a result of these four factors, tranche prices began to drop in early 2007, the market having lost faith in the subprime mortgage loans underlying many structured financings. Two Bear Stearns hedge funds collapsed. Everything that could possibly go wrong then did.

We are now light-years, if not parsecs, away from the Three C’s of Lending. Banking is a relatively simple business. These are the three principles of lending - Character: Is the borrower good to his promise to pay you back? Capacity: Will the borrower, in his business, have enough cash flow to repay the loan. Collateral: If all else fails, is the loan backed by sufficient collateral and/or guarantees? Bankers from times immemorial have more-or-less followed these principles ⁶. There have, of course, been banking crises ⁷ ; but the Crisis of 2008 is unique due to the involvement of structured finance and therefore its world-wide scale.

The Financial System

The role of the financial system is to direct savings to their best economic uses. We described the statistical abstractions of structured finance in some detail to show they had very little to do with this task (at best they did not impede sound lending); however later supplying the ideology and funding for a binge (the 3C’s of lending having been ignored).

Current discussions for financial system reform suggest salary caps, improved incentives, and regulatory restructuring. We think none of these are likely to be effective. More important is the evolution of the financial system itself. In “Financial Darwinism,” consultant Leo Tilman writes:

The need for dynamic management of business models of financial institutions and investment portfolios is self-evident. Due to secular trends, an increasing proportion of economic performance of financial institutions and institutional investors is likely to be driven by active risk taking as opposed to market-insensitive fees or balance sheet arbitrage. Therefore economic peformance (sic) is likely to become increasingly cyclical, with earnings and returns progressively at the mercy of market environments.

…In periods of lower returns and compressed risk premia, financial institutions and institutional investors have to choose between (a) accepting lower returns and earnings, (b) employing increased leverage and more complex products while maintaining static business models, or (c) using dynamic management and business model transformation to combat margin and earnings pressures. ⁸

In other words, the performance pressures on financial institutions are intense; and the profitabilities of the basic businesses of lending and fee based investment banking are limited. Due to the impact of the public markets upon lending spreads, banks really don’t like wholesale commercial lending because a prime rate loan is usually a break-even proposition after the cost of funds and overheads. Large commercial banks would rather earn fees by securitizing loans and taking proprietary risks with their market investments, like Goldman Sachs. They therefore lost their main function of directing savings to their best uses, that function now carried out by Mr. Market. But there are things the market doesn’t do well, like being accurate when billions of dollars are at stake or taking a long-term investment view ⁹, a time frame required by productive processes like manufacturing.

Most crucially, can the complex process of dynamic risk taking be managed? It takes exceptional skills. In “The Road to Financial Reformation,” Henry Kaufman describes how mega-financial institutions are managed:

Good governance in financial institutions – or any organization for that matter – begins at the top, with the integrity and skills of the leaders. Easy as this may sound, it is extremely difficult to attain within large institutions…Senior management…must depend on middle management for a flow of accurate information, but also must rely on the skills of subordinates in modeling risks, across a highly diverse range of activities ¹⁰.

But consider now what happened to AIG, a too big to fail financial institution that the government had to rescue. In 1986 Howard Sosin, from Bell Labs, convinced Hank Greenberg, CEO of AIG, to lend the company’s AAA credit rating to a new quantitative finance venture that would do long-term interest rate swaps. What evolved, according to the 12/29/08 Washington Post was, “…a culture of skepticism. The firm set up a committee to examine all transactions at the end of each workday, searching for flaws in logic, pricing, and hedges…Sosin and his colleagues worked to create a finely balanced system that married technology, intelligence, verve and cultural discipline.”

However, there were changes. In disagreements with Greenberg, most of the firm’s founders left. Joseph Cassano, who processed transactions without a background in risk management and hedging, became the nominal and then actual CEO of the group. According to the Post, “Cassano emerged as Greenberg’s candidate to take over. Some colleagues questioned his qualifications to manage a team that was heavily dependent on quantitative skills. Though he was the firm’s chief operating officer, some colleagues thought that he wasn’t conversant with the complex calculations of risk that remained at the heart of the business.” Under Cassano, AIG grew the unhedgable credit default swap (insurance) business, exposing the company to more than $500 billion in liabilities and to contract provisions that called for AIG to provide more collateral as the company’s credit sank. As the credit markets imploded, the unthinkable because actual; and AIG then had to be rescued by the government.

As the reader will gather, quantitative finance is the province of PhDs in financial economics. It belongs at hedge funds and specialized companies, and should not impact the credit-generating banking system where it can’t be properly managed. We return to the reason why finance exists: to direct savings to their best economic uses. Present reform discussions in Washington are but incremental:

1) Make the Fed responsible for systematic regulation, although it did not foresee the current crisis.

2) Create a new independent regulator, although the one in England also failed. The new regulator would likely not have the independence to effect the wide reforms that would be necessary.

3) Align the long-term financial interests of executives with their shareholders, although many at Lehman and Bear Stearns had large portions of their net worths tied up in company stock.

Current discussions about financial system reform either aim at improving the present system of administrative discretion, or are structural. In Federalist No. 51, James Madison wrote, “If men were angels, no government would be necessary. If angels were to govern men, neither external nor internal controls on government would be necessary.” ¹¹ Financial system reform should not depend mainly upon the people appointed to government, now and in the future. The more robust reform would be structural, splitting the lending functions of banks from their more imaginative market-based businesses. If this sounds suspiciously like Glass-Steagall, you are right.

The Glass-Steagall Act of 1933 separated commercial from investment banking. Passed in the wake of the bank underwriting abuses of the era, it prevented the large-scale failure of the U.S. banking system until it was abolished in 1999.

England has a market-based banking system similar to that of the U.S., and there are similar discussions how it should be reformed. The Financial Services Authority, its regulator, argues against a Glass-Steagall like reform. Banking industry arguments in the U.S. are similar:

The practical question, however, is whether it is possible to draw a clear legal distinction between allowed and prohibited activities in the modern economy of increasingly blurred boundaries among asset classes (our note: modern finance is atomic finance), floating exchange rates, global capital flows and a significant role for securitized credit. ¹²

But to again note, the financial industry exists to direct savings to its best uses. The industry’s opaque complexity is a problem, not a reason against its restructure. The major business argument against dividing the financial industry into public utility banking and all other activities is less its complexity than the lack of profitability of wholesale commercial bank lending, in short-term competition against the capital markets. Under normal conditions, large companies can raise funds more cheaply in the capital markets than to go to the banks. Put simply, the situation from a societal standpoint is, pay now with slightly higher interest rates; or pay later when the world-wide market melts down, as volatile markets will always do. A solution to this dilemma has to be institutional. We do not comment further on this other than to point out that the costs of this crisis to the U.S. will be in the trillions of dollars.

The Administration’s approach to banking industry reform has so far been incremental. It does not match the scope of the problem. Paul Volker has advocated breaking up the giants, turning them into lenders, and “The government in return would rescue banks that fail,” according to a 10/21/09 NYT article. This is structurally the best proposal, but the issue of banking industry profitability should also be addressed.

In a 10/21/09 letter to the NYT, John S. Reed the former chairman of Citigroup wrote, “…some kind of separation between institutions that deal primarily in the capital markets and those involved in more traditional deposit-taking and working-capital finance makes sense.” Such a separation would improve lending and allow focus upon the real issues facing the economy: exports (that is relevance to the rest of the world), jobs, and innovation.

Footnotes

__

Glass-Steagall would have prevented commercial banks from underwriting, and possibly holding (by proprietary trading and investing), the toxic assets that almost crashed the financial system. Paul Krugman says that the general goal of reform should be to “reduce bankers’ incentives and ability to concentrate risk….” Citigroup, Bank of America, and Wachovia were major originators of the super-senior tranches of CDOs (NYT, 7/24/07), securities that concentrated rather than dispersed financial risk. Because the commercial and investment banks (notably Merrill Lynch) were able to convince AIG and other insurance companies to write credit default swaps on these tranches, the banks thought they were investing in AAA securities. Thus, as related by Gillian Tett:

One clue to what had gone so terribly wrong at Citi could be found in the dry details of the technical statement that Citi issued that day and then filed with the Securities and Exchange Commission. The statement noted that Citi had $85 billion of exposure to the subprime market sitting on its books of which $43 billion* was in the form of super-senior risk attached to…(CDOs). That was a staggeringly large number, made doubly shocking because it had never been highlighted in any report that Citi had previously issued. Out of the $43 billion, around $18 billion was linked to investment-grade and mezzanine CDO of ABS, which Citi had retained on its balance sheet, either because it was in the process of creating CDO or because it had nowhere else to park it as it cranked up its CDO machine back in 2006 and 2007 (our emphasis). Citi’s exposure to the strike in the commercial paper market was also staggering. Another $25 billion of the bank’s total exposure was “commercial paper principally secured by super-senior tranches of high-grade ABS CDOs…”

(Tett quotes a senior Citigroup banker) “Perhaps there were a dozen people in the bank who really understood all this before – I doubt it was more.”

Gillian Tett

Fool’s Gold (2009)

p.p. 204-206

* Citigroup’s book net worth on 12/31/07 was $113.4 billion, $72.4 if you deduct goodwill.

__

What happened at Citigroup? In 4/8/10 testimony before the Financial Crisis Inquiry Commission Robert Rubin, former chairman of the company’s executive committee, testified that there was no failure in the bank’s system of risk management. The problem was that the bank believed the AAA ratings that the ratings agencies gave the subprime CDOs, resulting in $30 billion in losses that almost crashed the bank. Citigroup considered these securities more than prime credits. We note: these investments, many buried in the one trillion dollar trading book, were held by leverage. Thus any failure in these investments – although a small percentage of the total - would create a massive problem for the bank and for the financial system.

This presentation indicates that the bank had a structured finance credit research department in 2006, a year before the market melted down. Statistical models are ordinarily continuous. The authors describe CDO pricing in terms of three Poisson jump processes, whose parameters are tuned using historic data from 2003-2005 (sic). We had taken one look at the parameters of the Poisson statistical model and decided that complicated quantitative finance was generally inapplicable (at least for a MBA) to the financial markets. We do not try to reconcile the first paragraph with the second.

__

The Financial Crisis of 2008 was caused by misapplied financial models and, simply, really bad lending that had no connection with those models. The 9/27/10 NYT relates:

As the mortgage market grew frothy in 2006 – leading to a housing bubble that nearly brought down the banking system two years later – ratings agencies charged with assessing risk in mortgage pools dismissed conclusive evidence that many of the loans were dubious, according to testimony given last week to the Financial Crisis Inquiry Commission….

D. Keith Johnson, a former president of Clayton holdings, a company that analyzed mortgage pools for the Wall Street firms that sold them, told the commission…that almost half the mortgages Clayton sampled from the beginning of 2006 through June 2007 failed to meet crucial quality benchmarks that banks had promised to investors.

Yet, Clayton found, Wall Street was placing many of the troubled loans into bundles known as mortgage securities.

Mr. Johnson said he took this data to officials at Standard & Poor’s, Fitch Ratings and to the executive team at Moody’s Investors Service.

“We went to the ratings agencies and said ‘Wouldn’t this information be great for you to have as you assign tranche levels of risk?”’ But none of the agencies took him up on his offer, he said, indicating that it was against their business interests to be too critical of Wall Street.

“If any one of them would have adopted it,” he testified, ‘they would have lost market share.”

In a mania, people lose their perspective and don’t want to hear the bad news. Also, when the numbers get large, people become less careful; add to that short-term bonus incentives.

__

A $1 billion mortgage fraud lawsuit filed by the U.S. District Attorney on 5/3/11 against Deutsche Bank illustrates how bad residential real estate lending had become. In this significant instance, it also illustrates that that the government did not displace a virtuous private sector, as the Republican right claims. It illustrates that MortgageIT, a subsidiary DB acquired in 2007, took full advantage of government inattention to obtain the FHA guarantees that would enable DB to securitize more mortgages and increase its fees. At this writing, DB has yet to reply to this complaint.

Complaint: Between 1999 and 2009, MortgageIT falsely certified and represented their compliance with HUD quality control requirements, permitting MortgageIT to endorse more than 39,000 mortgages for FHA insurance (totaling more than $5 billion in principal obligations). “As of February 2011, HUD has paid more than $386 million in FHA insurance claims and related costs arising out of Defendants’ approval for mortgages for FHA insurance….The government expects HUD will be required to pay hundreds of millions of dollars in additional FHA insurance claims as additional mortgages underwritten by MortgageIT default in the months and years ahead.”

The Program: The Department of Housing and Urban Development’s Federal Housing Administration is the world’s largest mortgage insurer, “(The Direct Endorsement Lender Program encourages) lenders to make loans to borrowers who might not be able to meet conventional underwriting requirements.” The Program requires lenders to implement a quality control plan that, among other things, verifies: 1) The adequacy of the borrower’s income to meet mortgage payments and other obligations, 2) The borrower’s record of creditworthiness, 3) The valuation of the property. The lender then approves qualifying mortgages, and obtains a mortgage insurance certificate from the FHA that relies upon these lender representations.

Allegations: The lawsuit alleges that “DB and MortgageIT repeatedly lied (our emphasis) to HUD to obtain and maintain MortgageIT’s Direct Endorsement Lender status. DB and MortgageIT failed to implement the quality control procedures required by HUD, and their violations of HUD rules were egregious.” The complaint cites the following specific default examples, and the major violations of HUD procedure:

1) The Center Street Property, New York (loan funded 2002)

“…MortgageIT endorsed the Center Street Mortgage Application without proof that the borrower closed with gift funds (downpayment) from a proper source rather than from, for instance, the seller.”

2) The Bittercreed Drive Property, Colorado (loan funded 2004)

“…MortgageIT failed to develop a credit history by assembling any such records in reviewing the Bittercreed Drive Mortgage Application, even though the borrower had no established credit history…”

3) The Monument Avenue Property, Indiana (loan funded 2005)

“…MortgageIT failed to verify and document the borrower’s purported investment in the…Property; indeed, the documentation…reveals that the borrower had documented assets of thousands of dollars less than the amount the borrower was purportedly investing in the property.”

4) The Kentucky Street Property, Michigan (loan funded 2005)

“…MortgageIT failed to contact the employer and, after the mortgage closed, the listed employer verified that the borrower was never its employee.”

Our Analysis: Private enterprise was responsible for the bad lending and simply took advantage of government inattention.

1) “…a HUD audit conducted during the week of September 13, 2003 by the HUD Quality Assurance Division, Philadelphia Homeownership Center, revealed that MortgageIT had ‘not maintained a Quality Control Plan …in accordance with HUD/FHA requirements,’ and that, among other failures, MortgageIT had failed to ‘ensure that loans that go into default within the first 6 months are reviewed.’…MortgageIT responded to the 2003 audit by informing HUD that it had altered its quality control procedures to follow HUD rules, including by ensuring the review of all early payment defaults. That representation was false.” (HUD should have verified compliance at the time.)

2) “Until late 2005, MortgageIT had no personnel to conduct the required quality control reviews for closed FHA-insured loans. In or about 2004, Mortgage IT contracted with an outside vendor, Tena Companies, Inc. (‘Tena’), to conduct quality control reviews of closed FHA-insured loans.…Throughout 2004 Tena prepared findings letters detailing underwriting violations it found…No one at MortgageIT read any of the Tena findings letters as they arrived in 2004. Instead, MortgageIT employees stuffed the letters, unopened and unread, in a closet at MortgageIT’s Manhattan headquarters.”

HUD failed to detect that there was no effective credit function at MortgageIT. We do not know if HUD procedures required an annual program review of the lender, but there was apparently no means of followup or enforcement. The responsible person between 2001-2008 at HUD was appointed by the Bush administration.

_

In 2013, a $5 billion lawsuit was finally filed against Standard and Poor’s, the world's largest credit rating agency.

Complaint:

A lawsuit filed by the U.S. District Attorney on 2/4/13 in California against the rating agency, Standard & Poor’s, alleged that between 2004-2007, “…S&P, knowingly and with the intent to defraud, devised, participated in, and executed a scheme to defraud investors in RMBS (our notes: residential mortgage backed securities) and CDO (collateralized debt obligation) tranches, including federally insured institutions…falsely (representing) that its credit ratings of RMBS and CDO tranches were objective, independent, uninfluenced by any conflicts of interest that might compromise S&P’s analytic judgment, and reflected S&P’s true current opinion regarding the credit risks the rated RMBS and CDO tranches posed to investors.”

The Extent:

1) Between 2004 and 2007, S&P issued credit ratings on over $2.8 trillion worth of RMBS and nearly $1.2 trillion worth of CDOs.

2) In 2007 Global CDO generated revenues of approximately $203 million for the company.

3) In 2007 Global ABS generated revenues of more than $243 million.

Allegations:

To protect and increase its market share of the business that came from the (then) investment banks like Merrill Lynch and Bear Stearns, the company:

1) Chose the financial model giving the highest ratings. “If transaction fails E3, then use E3 Low.”

2) Bent the Gaussian Copula model to suit business needs. “In 2006, at a meeting…S&P loosened to zero its correlation assumptions (a key measure of default risk) between 'a CDO of ABS asset' and 'an RMBS asset in a CDO/ABS transaction'…business personnel were in favor of this decision which was made without the benefit of any data (our notes) and would lead to S&P’s rating models arriving at lower estimates of credit risks for CDOs collateralized by such assets. Commenting on this change on or about April 2, 2007, a CDO analyst indicated to a former coworker that it resulted in a loophole in S&P’s rating model big enough to drive a Mack truck through.”

“…CDO team leader…clearly knew…is ultimately responsible.”

3) Covered up increasing defaults in subprime RMBS, “…because of concern that S&Ps ratings business would be affected if there were severe downgrades.”

Our Analysis:

S&P has a real problem. The long chain that developed, from mortgage origination to funding, diffused responsibility. But the credit agencies were responsible for the final overall determination of investment default risk. Their favorable ratings enabled these structured mortgage financings to be distributed throughout the world’s financial system, bringing it to the brink of collapse.

RETURN TO PREVIOUS PAGE