A new discussion paper from the European Central Bank and the Bank of England, “The case for a better functioning securitisation market in the European Union,” has introduced the idea of a “qualifying securitisation” as an instrument for promoting the liquidity of credit markets. Its an interesting idea, but our first thought is, is this just a nice way of not calling something a Structured Investment Vehicle or a Special Purpose Vehicle?
The idea of a qualifying securitisation is basically to put a legal structure about a secured product that commoditizes the framework for the benefit of risk analysis. “The view of the Bank of England and the ECB is that a ‘qualifying securitisation’ should be defined as a security where risk and pay-offs can be consistently and predictably understood.” These products wouldn’t be risk-free but the Banks are suggesting that they would be lower risk than any other securitisation, and that this should create a positive bump for liquidity.
Maybe there would be some Basel III capital benefits as well. The Banks note that two benefits of qualifying securitisations could be: “Improved liquidity and reduced credit risk of the securitisation may justify a different regulatory capital and liquidity treatment for some or all of its tranches; and to the extent that the risk characteristics of some assets are successfully improved, these may naturally be reflected in haircuts for central bank liquidity operations that already accept securitisations as collateral.” This is like the use of CCPs for OTC derivatives or any other product. The risk is still there and its all still the same counterparties (clearing members), but the products are treated differently for capital purposes to achieve some other goal.
Joseph Coterill, writing in the Financial Times, noted that for all its good ideas, the fact is that the collateral analytics part of securitised products still doesn’t get attended to well enough: “producing yet further reams of legal documentation (to create a new over-class of ‘qualifying securitisations’) might be bad for risk management generally. Issuers could hide loopholes in plain sight.” We agree.
Securitisation provides a legitimate service by aggregating and diversifying credit risk to interested investors while getting capital to companies that need it. We do a lot of work on collateral analytics, and the fact remains that the collateral underlying any securitized product can be high quality or not. This information is not always captured by ratings agencies and it is incumbent on any investor to know what they have and what they don’t. Calling something “qualifying” doesn’t decrease the risk of the underlying collateral, even when (and sometimes especially when) it starts out as a basket that is supposed to look pretty homogeneous.
We like the idea of qualifying securitisations, but let’s be serious: no matter the wrapper, the safety of these products are based on what backs them. If the collateral is good, then great. If not, then there will be trouble. We’d like to see the next step of this discussion paper get to how investors will track and monitor the assets that underlying qualifying securitisations as a better step in risk management, and get fairly compensated for that risk.
4 Comments. Leave new
I wouldn’t bundle SIVs and SPVs together so bluntly. Maybe it’s just semantics, but what made pre-crisis SIVs (and some ABCP conduits) dangerous was that they were essentially resecuritization vehicles designed to game credit rating methodologies and risk management models. True you could say that about some other securitization vehicles (e.g., CDO-squareds) but I think we should be careful not to paint everything with the same brush.
I’ll make the counterargument that calling something a Qualifying Securitisation is effectively setting up a new game to play. I agree that not all SIVs, SPVs etc are equal at all, but the truth to that is in the underlying collateral, not the inherent name. Perhaps Qualifying Securitisations would be different?
I hope so! I’m encouraged that Box 3 of the BoE/ECB paper seems to imply the exclusion of resecuritizations and features that are designed to game risk models and just generally obfuscate risks (see paragraphs 125 and 128).
I’ll take a more cautious wait-and-see approach before I get overly encouraged.