Editor’s Note: in this guest post, former bond insurance insider Steve Ruterman discusses important considerations for investing in private label MBS beyond credit risk analysis, including how investors can benefit from understanding the differences in servicer behavior and business models. Such analysis can be combined with a strategy of selecting bonds that have the potential for significant upside from the enforcement of creditor rights, such as servicer termination and/or loan put-backs, to generate even greater returns for fixed income investors. Those who wish to learn more about how to take advantage of either of these types of analyses with respect to MBS investments are welcome to reach out to me or Mr. Ruterman – IMG.
By Steve Ruterman, guest blogger
Speaking as a major consumer of televised nature programs, it has been a lot of fun to watch the ongoing cyclical migration of yield-hungry fixed income investors. Just now, they are approaching subprime mortgage backed securities, which they know to have been dangerous in the past (see here, for example). They are doing so because they believe the past danger came from the terrible credit quality of the loans in the pool, and that such risk may be exaggerated in the current pricing of the bonds.
Keep in mind that 2005 vintage pools have almost seven years of seasoning, so the downside risk of future defaults must have been at least somewhat mitigated. Seasoning used to be considered a good thing: prices are low, and yields are attractive. Other members of the ecosystem are now crowding around the pools. We live in a competitive world, after all, so why not jump in?
My answer to the question is, “Jump in if you must, but first take the trouble to understand what kinds of critters also live in these pools.”
I refer, of course, to the loan servicers and the risks they may represent to investors. These risks are independent of loan credit quality, and are often discounted by fixed income investors, despite their potentially significant impact.
I’m sure that readers of The Subprime Shakeout can readily accept the proposition that some loan servicers are better or worse than others. So, how might an investor go about understanding those differences, and how might they reflect them in their valuations and projections of risk adjusted returns on RMBS?
In theory, it should be possible to run some objective analytics and conclude which servicers are doing a better job for bondholders than others. A simple comparison of prepayment speeds, delinquency rates, and net credit loss experience across several servicers should do the trick, particularly if the analyst compares like asset classes and vintage origination years.
However, as previously discussed here, servicers may be incentivized to misreport speeds, delinquency rates, and/or losses. Examples include:
- A widespread industry practice in effect at present actually results in increased unpaid loan balances due to the modification of loan terms. This is because the purpose of the modification is to claw back servicer advances on an accelerated basis. Since servicers do not always report the volume of claw-back modifications monthly, there are no simple means for adjusting reported speeds.
- The reported delinquency status of loans in loss mitigation or modification queues is flexible and can really be whatever the servicer wants it to be.
- Some second lien mortgage servicers are infamous for failing to charge off loans more than six months delinquent although they are generally contractually obligated to do so (and are now required to do so by the terms of the Attorney General Foreclosure Settlement (“AGFS”)).
These are just a few examples of known servicer reporting issues which can materially affect the investor’s analytic efforts.
As in the last subprime mortgage cycle, many subprime loan pools have new servicers that have replaced the originals (see Moody’s comment). Even if the same servicers are still in place, recent increases in servicing costs and regulatory pressures (such as the AGFS) have forced changes in old business models. Some original and replacement loan servicers have their own agendas, and in the pursuit of their own perceived best interests, they often act against the best interests of investors. In fact, some servicers often take actions which cause dollar-for-dollar losses to the investors that rely on them.
The key to assessing the relative risk to an investor’s returns from any loan servicer is assessing the loan servicer’s business model. The investor always wants the servicer’s business model to achieve the following goals:
- Aggressively prevent current loans from rolling into early stage delinquency.
- If loans do roll into delinquency, take all necessary steps to cure the delinquency.
- Should any loans roll into the late stages of delinquency, take all steps necessary to mitigate potential losses and cure as many as possible.
- If all else fails and foreclosure is necessary and desirable in order to protect positive loan values, act quickly to minimize foreclosure and REO timelines.
Sounds simple, right? Unfortunately, this recipe does not describe the business model of some loan servicers, who spend as little money as possible on 1) and 2), above, for example, because they prefer to maximize the number of delinquent borrowers paying them late fees.
If the current disastrous state of the mortgage markets has anything to teach us, it is this: investors are going to have to do more to protect themselves than they have been accustomed to doing [Editor's note: and they won't be able to rely on the AGFS for meaningful change in servicer behavior -IMG]. Risks from servicers’ business models are quite material, and can only be avoided by understanding what each servicer’s model is, and is not.
Steve Ruterman is an independent consultant to institutions and institutional investors with significant RMBS exposures and a fan of The Subprime Shakeout. He recently retired after a 14 year career with MBIA Insurance Corporation, during which he transferred over 20 mortgage loan pools to new servicers. Mr. Ruterman welcomes your comments, and can be reached by email at Steve.Ruterman@yahoo.com.