CMBS 2.0 is coming, we hope (and pray). But boy, it’s taking its good time about it. Putting aside what our friends in Washington may or may not do to the structure of securitization, it’s remarkable to me how shy we in the industry (and its trade organizations) seem to be about putting a stake in the ground as to what CMBS 2.0 should look like.
With CMBS 1.0, we built the airplane while flying it, so it's hardly shocking that when tested, some things failed the stress test. On the other hand, we also did a great deal of fundamental work on an industry-wide basis in the early days, to make CMBS work. We created the IRP, the data dictionary and the like. Shouldn’t we do at least that much again?
Now that we’ve had a chance to observe the problems of CMBS 1.0 in the crucible of a wrenching recession, we seem mildly disinclined to take any dramatic action to address structural problems on an industry wide basis.
There are lots of good ideas floating around (ok, some really bad ones as well). The servicing community, the investor community and the Street have all weighed in, but there doesn’t seem to be much energy around making real, industry-wide, best practice fixes. We’ve seen four deals that, more or less, constitute CMBS 2.0 with a range of new features. But, on balance, a fair minded observer would have to conclude it’s all a tepid response to the existential failures of the current industry model.
Why not embrace best practices to address those things which seem ripe for a fix? Things such as acknowledged problems in the waterfall of reimbursing accrued interest ahead of principal recovery, the exclusion of appraisal reduction from the control class change mechanic, the variability in servicing standard, the absence of clarity about the discount rate for valuing specially serviced loans. Let’s also look at data, at transparency and surveillance.
Why not (gasp!) try to standardize, at least, some of the PSAs. Many of the key components could practically be standardized without enormous brain damage. So underwriters want to brand their product with a particular form of PSA, fine, but there is no reason why we ought, for example, to have multiple definitions of the servicing standard. There’s no reason why we can’t find key provisions in roughly the same places across program PSAs.
Finally, can’t we stop for a moment and make the disclosure documents and PSAs more readable, more narratively clear and understandable? These documents have grown up like coral reefs from original precedent. Genetic drift has not, broadly, been productive. We have a chance to prune the wild things back into something more coherent, shorter and more understandable. It would be good for the market for CMBS 2.0 relaunch and for liquidity.
The counterpoint, of course, is that investors really won’t pony up for better documents, better disclosure, more data, etc. Someone will do all the best practice things championed by the various IG constituencies and see no premium. Maybe that’s the case but it’s still no reason not to fix what’s broken. At the end of the day, the fixes will still have a positive difference in the new cycle. OK. If I was a bettin’ man, I’d bet we’d not do these things, but we should.