Voltron's economics blog. Started in Iraq in 2007 as the "Gamblers Anonymous Support Group" email list.
Monday, March 31, 2008
The Spotlight's On U.S. Alt-A RMBS Issuers As Performance Deteriorates Rapidly
Voltron says: great data in this article. Take a look at table 1. Lehman is right behind Countrywide and Bear Stearns.
The underperformance of mortgage loans backing U.S. residential mortgage-backed securities (RMBS) at first seemed limited to the subprime mortgage sector, but has since filtered through to virtually every corner of the non-prime mortgage industry. The Alt-A market is no exception.
"Alternative-A" loans are first-lien residential mortgages that generally conform to traditional "prime" credit guidelines, although the loan-to-value (LTV) ratio, loan documentation, occupancy status, property type, or other factors cause the loan not to qualify under standard underwriting programs.
Those underwriting guidelines began to loosen in late 2005, a trend that gained momentum in 2006 as Alt-A originators and investors became more accepting of so-called "layered risk" (the inclusion of multiple high-risk attributes within a single loan). The proliferation of layered risk, along with stagnant or declining home price appreciation (HPA), are factors that have led to the current deterioration in performance. Now, because underwriting standards have tightened in 2007 and HPA continues to fall, leveraged Alt-A homeowners who run into financial trouble have fewer options to refinance their loans, and many are finding it difficult to avoid foreclosure.
In a continued attempt to add transparency and insight into the U.S. residential mortgage markets, Standard & Poor's Ratings Services will release a series of articles commenting on U.S. RMBS collateral performance. In this article, we break down the relative performance of the top 20 Alt-A issuers and look at the market in general, which has seen a sharp performance deterioration in the 2006 and 2007 vintages. Future articles will provide a detailed analysis of underwriting and risk management practices, collateral trends, and economic factors that are contributing to the market downturn, both in the Alt-A sector as well as within other segments of the U.S. RMBS market.
Delinquencies Are High For 2006 And 2007 Alt-A Vintages
Alt-A severe delinquencies (90 days or more, including loans in real estate owned, or "REO," foreclosure, or bankruptcy) have been increasing in recent months, with the number of delinquencies in the 2006 vintage rising to more than double those of the 2005 vintage and more than four times those of the 2004 and 2003 vintages (see chart 1). Both the speed and magnitude of the 2006 vintage performance deterioration is anomalous, given similar loan seasoning compared with prior vintages. What's more, the negative performance trend doesn't seem to be abating for the 2007 vintage. While it's too soon to draw a conclusion regarding the ultimate 2007-vintage performance, early loan seasoning suggests that the deteriorating trend may lead to the worst-ever loss performance within the Alt-A market (see chart 1).
Still, it's important to note that in spite of the recent spike in delinquencies, the vast majority (about 90%) of Alt-A homeowners are current on their mortgage payments. Moreover, while early-period delinquencies have been higher than expected, actual cumulative losses to date remain relatively low. It will likely take an extended period of high delinquencies (and ultimate losses) to pierce investment-grade credit enhancement levels across the overall market, although individual deal performance will vary. Chart 2 demonstrates this point by comparing cumulative losses to date against average 'B' credit enhancement levels. For the most seasoned 2002 vintage, cumulative losses to date are just over 50% of the original 'B' credit enhancement levels. (Note: investment-grade or 'BBB' credit enhancement levels are much higher at 1.26% for the 2003 vintage, 1.72% for the 2004 vintage, 2.08% for the 2005 vintage, and 2.57% for the 2006 vintage.)
Issuer-Specific Delinquencies Vary Widely
The Alt-A market is not homogeneous, but has a diverse mix of borrower profiles and product types. As a result, our opinion of the level of credit risk in the market can vary greatly from issuer to issuer, leading to a wide range of expected losses. The higher end of the Alt-A market (high FICO scores, few layers of risk) closely parallels the prime jumbo market and, therefore, is likely to experience closer to "prime-like" performance. The lower end of the Alt-A market (low FICO scores, many layers of risk) is very different and is concentrated on loans with many compounding risk factors, so loan delinquencies here should more closely follow "subprime-like" performance.
In the below tables and charts, we highlight 2006 vintage performance for the top 20 Alt-A issuers, which collectively represent almost 95% of the total Alt-A market. Severe delinquencies for the Alt-A 2006 vintage are 4.61% (as of September, the latest data period available in the LoanPerformance database). However, delinquencies for the top 20 issuers vary from 2% to almost 10%, a notably wide range (see table 1 and chart 3). Given the diversity of the Alt-A market, we expect issuer performance to vary based on an issuer's strategic position within the market, the types of product and credit risks it favors, and the sophistication and discipline of its risk management, underwriting, and quality control.
Standard & Poor's 2006 Rated Issuance (Top 20)
Rated issuance (bil. $)
Severe delquencies (%)
Avg. 'BBB' S&P credit enhancement (%)
Bank Of America
Clearly, collateral quality plays a fundamental role in the diverging performance trends. We expect issuers with a concentration of high-quality loan characteristics (such as high FICO scores and low combined LTV (CLTV) ratios) to experience better performance than those with lower-quality characteristics. Table 2 shows FICO score, CLTV, and average 'BBB' credit enhancement for the issuers with the highest and lowest delinquency performance. Although the data does show that collateral mix is correlated with performance, it would take a far deeper analysis to fully tease out variations in collateral quality. There are many other loan attributes (in addition to FICO and CLTV) that contribute to a particular loan's risk profile and that ultimately impact performance.
Product type introduces another strong bias in performance. Unique product risks inherent in certain loans, such as payment option ARM (POA), hybrid ARM, and fixed-rate loans, will lead to substantially differing performance over time (see charts 4-6). The divergent borrow profiles, payment shock characteristics, amortization schedules, and other loan attributes across products lead to a wide range of expectations for the timing and severity of ultimate losses.
Of the three product types highlighted below, fixed-rate loans have the lowest average delinquency levels of 3.47% (see chart 4). This is expected. The stability afforded by the fixed interest rate eliminates all payment shock associated with adjusting rates. This clearly lowers the risk profile of the product over the entire life of the loan. However, the structural protection is a medium- and long-term risk mitigant and is unlikely to be playing a direct role in the low early-period delinquencies that we are seeing in the 2006 vintage. The high credit quality of the typical fixed-rate borrower is likely having a greater impact on low initial delinquencies to date. Because the monthly payment of a fixed-rate loan is somewhat higher than that of other affordability products (hybrid ARM and POA loans), homeowners with fixed-rate loans tend to be more financially conservative and, consequently, migrate toward the structural benefits of the product.
Hybrid ARM products don't enjoy the same level of interest rate stability or conservative borrower profile, and, consequently, current severe delinquencies are at a much higher level of 6.26% (see chart 5). In general, borrowers will be subject to a "payment shock" once interest rates begin to adjust. (Note: there is a wide range of hybrid ARM products, from those with initial fixed-rate periods of two years to 10 years.) However, even for the shortest hybrid ARM products, payment shock has not yet emerged as a key stress on borrowers within the 2006 vintage; most Alt-A homeowners have many months remaining on their initial fixed-interest-rate period. But the self-selection of borrowers with weaker credit profiles into hybrid ARMs and other affordability products is affecting the current spike in severe delinquencies. Many of these borrowers have a financial need for the low initial monthly payments offered by hybrid ARM loans and may not be able to afford traditional fully indexed (or fully amortizing) rates. A hybrid ARM homeowner is presumably more financially stretched than a fixed-rate homeowner and, not surprisingly, is more likely to be falling behind on his or her mortgage payments, especially in light of falling HPA. The shorter the hybrid ARM period, the more vulnerable borrowers may be to rising monthly payments or falling HPA.
Finally, POA loans have the most product risk and, ultimately, are expected to experience the highest level of losses. Current average delinquency levels of 3.72% (see chart 6) are artificially low as a result of "teaser" payment options, which make it easier for homeowners to make their monthly payments during the first few years. The minimum required payment during the first few years of a POA loan is often only a fraction of the fully amortized amount. The vast majority of POA homeowners (more than 75%) have been opting to make only these minimum payments, resulting in a low performance hurdle for borrowers and generally leading to very few early-period delinquencies. However, we expect POA delinquencies (and losses) to rise sharply as loans reach their negative amortization ceiling, when borrowers are required to make significantly higher monthly payments (often two or more times the original payment). This "payment shock" is much more severe than that for hybrid ARM products and will occur at a time when homeowners have little (or no) equity built up in their homes (as a byproduct of negative amortization). While in the past POA borrowers may have been afforded ample opportunity to refinance, in the current market environment, troubled POA borrowers no longer have abundant liquidity available to them. Many borrowers in the future may find it difficult to avoid foreclosure.
Given these contrasting products risks, we feel it is important to view issuer-specific delinquencies by product type, which will provide a more relative view of performance.
Collateral Characteristics Don't Tell The Whole Story
Still, there are clear over- and underperformers within the market. Although subtle differences in collateral characteristics will go a long way in rationalizing diverging performance, they alone don't tell the whole story. In many cases, competitive advantages that issuers possess in underwriting, quality control, or risk management can serve as a key point of differentiation among the best and worst performers.
Ultimate performance will bear out these advantages over time. Given the importance of the issuer's role within securitizations, Standard & Poor's will adjust credit enhancement levels based on a particular issuer's strength or weakness in underwriting, risk management, quality control, or other due-diligence function. An upcoming commentary will provide a more detailed analysis of the various underwriting practices in the industry, including whether certain issuers remained more disciplined during the recent period of rapid growth, which may have allowed them to avoid significant exposure to the most troubled areas of the Alt-A market.
Note: Does not include Medicare, Medicaid or Social Security.
This is not an offer to buy or sell securities in any jurisdiction. This blog is for informational purposes only. I do not give legal, tax or accounting advice of any kind. I am not a licensed financial advisor. I make no representations as to the suitability of any transaction at any time. In fact, you should ignore everything I say.
Voltron's Current Forecast
Wall Street and the government are about to engage in battle and recriminations in the face of a massive wave of foreclosures. Wall Street will start yanking the leash, causing stock market crashes whenever the status quo is threatened.
The Fed's announcement that they are going to print a Trillion dollars and use it to buy debt is the proverbial "crossing of the Rubicon" The US Dollar is toast. I think the trigger will be some of the smaller holders of Treasurys (Singapore, for example) trying to dump their holdings before China does. This will cause a panic stampede out of Treasurys which will destroy the dollar.
Wells Fargo are liars. They have over 150 Billion in worthless home equity loans and a Trillion dollars in off-balance sheet liabilities.
Do not abuse leverage. The Fed is going to spend limitless amounts of money to pound interest rates and Gold prices into appearing "normal" when in fact they are on the verge of exploding. The tremendous volatility will shake out leveraged players.