The IRA is on the road this week, heading to DC to participate in the March 28th conference, "Mortgage Credit and Subprime Lending: Implications of a Deflating Bubble, " sponsored by American Enterprise Institute and Professional Risk Managers International Association, and do some business.

There is a lot of new activity in the regulatory sector regarding projects to build data collection capability focused on issues like the mortgage market, collateralized debt obligations or CDOs, and counterparty risk. Understanding who's got the ball in terms of concentrations of risk arising from the mortgage securitization sector has become a near obsession within certain US regulatory agencies -- and not a moment too soon.

The shadow currently hanging over the US mortgage industry is about liquidity, or the lack thereof. Both in the market for new loan originations all the way through to the investor in CDOs, the torrid flow of 2005 and 2006 is now a trickle. One fellow who understands the liquidity issue is Angelo Mozillo, the CEO of Countrywide Financial Corp (NYSE:CFC). He claims credit for sounding the alarm regarding subprime and asset quality generally a year ago. Mozillo very clearly is anticipating the demise of his less prudent competitors, as well as speculative investors in real estate.

CFC's lead unit reported a whole 4.8bps of aggregate loan defaults in 2006 vs. 16bp for the mortgage specialization peer group of the FDIC, a four-fold increase over 2005 for CFC but actually a decline for the peers. Mozillo does not need any VaR models or Monte Carlo simulations to tell him that the rest of 2007 is going to be ugly and he seems to relish the prospect. Part of the reason may be the strong financial profile the bank reported at the end of 2006, as shown below.

The IRA Bank Monitor: Countrywide Financial Corp -- December 2006

Source: FDIC, The IRA Bank Monitor

As liquidity is being taken away by the Fed and by the banks, Mozillo told Jim Cramer last week, there are "hundreds and hundreds" of subprime lenders being forced from business as the market rationalizes "below the radar screen." Happy days. The demise of these "irrational" lenders, as Mozillo rightly calls them, means a reduction in secondary market demand and the prospect of losses for investors in existing production.

Picking up market share by watching your competition destroy itself is not a business model for the faint of heart, but CFC has historically exceled avoiding the cyclical pitfalls of the mortgage market. Notice that the draconian Economic Capital model in the IRA Bank Monitor assigns a ratio of EC to Tier One Risk based capital of < 1 for CFC, 0.597, suggesting a relatively low-risk business model. Ditto the double-digit RAROC. In fact, because the CFC loan book has been so quiet for so long, most of its capital exposure in our EC model comes from securities risk.

CFC's lead bank currently is supported by capital equal to just 7.7% of total assets, almost a full standard deviation below peer. Based on quant models geared off of the past decade's default experience, CFC looks well capitalized, but Angelo better hope those retail loan default rates stay in single digits, just like the model says. In 2001, CFC reported 90bp of defaults, but has tracked < 5bp since then.

One telltale indication that Mozillo indeed knew the US mortgage bubble was a bust is that CFC raised its loan yield 150bp between 2005 and the end of 2006, this as his competitors were playing beggar thy neighbor, competing for the last marginal borrowers. Let's see if Mozillo has enough capital in CFC to ride the secular wave of rising mortgage defaults in 2007 and beyond.

Not everyone takes the sunny, pick the meat from their dead bones outlook of Mr. Mozillo on the US real estate sector. Many homeowners, lenders, traders and end-investors are burdened by huge amounts of risk shifted onto their shoulders by the Sell Side over the past decade. Nearly all involved share a common problem -- failing liquidity, falling prices in particular and widening credit spreads generally. Consider some examples we've heard in our travels:

The Home Owners: You are a two-career, "prime" couple in Washington DC who sold their townhouse in 2003 for $370,000 and bootstrapped into a $750,000 home in Potomac, MD, via an interest only mortgage. The mortgage balloons in a year. The home valuation, in theory at least, peaked over $1 million early in 2006. Today, with interest rates rising and DC quickly reverting back to a rental model for new construction, the two-career couple cannot find a bid for the house or a refinancing such that they could walk away at the close without writing a check.

The Manager: You are a mutual fund manager who purchased a number of CDOs over the past several years, including a series of deals brought in the 2005-2006 window when risk spreads were particularly tight and demand for collateral was fierce. Because of the high proportion of subprime paper in these deals, the effective bid for these specific CDOs has fallen dramatically as the ratings for the deals were downgraded. The fact that spreads have widened is also adding to the pain, which all told equates to roughly a 10% loss on the paper. Incidentally, the CDS for CDOs still trade significantly below the expected low double digit default rates on such portfolios. Go figure. Your auditor begins to ask questions when your prime broker balks at emailing a valuation for your portfolio.

The Lender: You are the CRO for a bank which sold some of its mortgage production during 2005-2006, but still has more than 45% of total assets concentrated in retail and commercial real estate lending. The bank has made good money originating, selling and even trading loans, and writing credit derivative insurance on CDO traunches. In fact, since 2003 the bank's portfolio of real estate loans has doubled as a percent of total assets. The bank's credit function has been satisfied with the risk profile, net long default risk and a hefty helping of duration as well, but the internal auditor just flagged the loans and CDS, and added an exception to a quarterly review, asking whether a valuation haircut need be applied to the bank's conduit and related derivatives. Fact is, the conduit is now effectively closed and the only bids in the market for loans purchased in the heady 2005-2006 window are well below par.

The Trader: You're a trader at a hedge fund which supplemented its returns over the past several years by writing CDS cover for CDOs brought to market by your prime clearing broker. The hedge fund's inventory of CDS contracts is financed by the prime broker. So far, the hedge fund has experienced no defaults covered by its CDS positions and took previous period premiums to income rather than creating a reserve for future contingencies. The indicated bid side of the five year CDS written by the fund against CDO deals has moved almost 1,000bp in the past 90 days and the prime broker is not willing to make a cash bid for older CDS positions. Given the change in pricing, the hedge fund's managers debate when and how to inform investors about the loss and, more important, the prospective future cost of performing on CDS obligations. The fund eventually suspends client withdrawals and reserves 30% of fund assets to cover CDS obligations.

All of these threads have one thing in common, reliance upon quantitative models employed by broker-dealers and the rating agencies to measure and price the risk of loans, CDOs and derivative securities. Pressure from both buyers and sellers of mortgage securitization deals helped boost collateral values and push credit spreads down to unrealistic levels, in the process causing a demand-driven surge in home prices which saw default rates fall to zero at many banks. At one point last year, CDS for subprime CDO deals was trading at less than 10% of the expected default rate on the collateral, like a "CCC" trading as a "A" credit default rating.

Over the past five years, the use of risk pricing tools such as VaR models and other types of statistical routines arguably amplified the effect of excess liquidity, boosting the throughput of the Wall Street mortgage origination machine, generating big fees, and vastly expanding the pool of risk for end-users. Now that process seems to be unwinding. For competitive souls like CFC CEO Angelo Mozillo, that fact seemingly spells opportunity, but only if loan default rates across the board stay below long-term averages.

Questions? Comments? [email protected]