I wrote about the disconnect between our CRE CLO technology and the task at hand (finding acceptable lever in an expanding leverage desert) in my last commentary. While the CRE CLO remains the best form of match-term, non-marked-to-market finance for portfolio lenders and represents the best alignment of interests between sponsor and investor across the capital markets, it’s a tool without a job now. The question raised last time, and again today, is: Can it be repurposed to do jobs that are in need of doing right now?
The last time I talked about multiseller transactions, I talked about breaking down silos between CRE debt and corporate debt and its potential use as a tool for disintermediating warehouse lenders from excess exposure.
I forgot something important: The use of CRE CLO technology as an alternative to the sale of CRE assets by lenders either needing capital relief or relief from excess concentration in CRE exposures.
Why not securitize as opposed to sell? Securitization can provide the same benefits of capital relief and general de-risking as an outright sale. Assuming it pencils, it can also avoid the potential further deterioration of the market value of the chosen pool (as well as the retained book) by avoiding exposure in an open market sales transaction. It also gives the lender an opportunity to put to work a considerable portion of the disposition/financing proceeds by investing in senior, generally investment grade, bonds.
Here’s how the transaction works. The lender would identify and craft a pool of assets for the transaction. This need not be exclusively the type of assets one would try to assemble to do a broadly syndicated CRE CLO (although that would be great). You certainly could include assets which are money-good but are criticized under the lender’s own risk rating system.
To make this transaction work, the lender is going to need an investor to buy the sub debt (Money Point: For my high-yield friends with a taste for it, this is going to be a major opportunity!). Depending upon the tax structure of the transaction, the investor will either need to be a partner or member in the issuer vehicle, or could simply be the sub debt buyer from the securitization. In this structure, the lender could retain the investment grade securities generated by the transaction and thereby put the sales proceeds or a portion of the sales proceeds back to work.
The customary structure for this type of thing is a CRE CLO with a Qualified REIT Subsidiary or QRS issuer. The QRS structure allows maximum flexibility in terms of managing assets and would allow the addition of additional collateral.
If a REIT is used, the REIT must hold the non-investment grade securities created by the QRS issuer as a matter of tax law. This would also achieve our goals of derecognition of the commercial real estate assets and a material reduction in risk-based capital.
In this structure, the investor could own the REIT outright or, alternatively, the investor could be a member of an upstream vehicle which owns the REIT with significant control over the business of the REIT and with a special allocation of the gains and loss of the retained non-investment grade securities. (The upstream vehicle could also retain some or all of the investment grade securities which would be for the account of the lender.) The investor could also be the risk retention party.
As an alternative to a REIT, the vehicle could be organized as a REMIC and this has certainly been done before. If organized as a REMIC, all of the related securities can be transferred to third parties in outright sale transactions, but there are, of course, certain operating limitations associated with REMIC status.
The third approach is to use a simplified capital structure with just a single senior and a single subordinate class. The notes would be senior/subordinate and not sequential, meaning cash would be distributed pro rata unless a default or another triggering event occurs, in which case the senior class would be turboed. Neither a REIT nor a REMIC election would be required yet the structure would not constitute a taxable mortgage pool. Note, however, that this nonsequential capital structure might have something of a negative impact on ratings levels and pricing efficiency.
In all three of these structures, the investor must have control because in order to derecognize the commercial mortgage loans and get the required risk-based capital pickup, the party with the predominance of control over the business of the entity will be the party which would consolidate the assets and liabilities of the vehicle for GAAP purposes. GAAP treatment here will drive RBC treatment.
Okay, I know that in the first instance, the paramount question, and in some ways the only question is, does this pencil? If the economics of pursuing this type of an exit strategy are wildly less attractive than an outright sale of the pool of loans, the structure might not work, or at least it would make the value proposition of this structure more uncertain.
But, in addition to a straightforward dollar and cents analysis (and remember that this calculus needs to be done on a net dollar basis as all of these alternatives will have a significant cost drag), there may be other factors at play here. There may be positive externalities that might materially weigh in favor of a structured solution.
What could the structure do for us?
- The structure would allow derecognition of the loans from a GAAP perspective.
- RBC would be reduced from 100% to a 20% charge only against the retained bonds. In connection with a full sellout of all the investment grade bonds, there would be obviously no risk-based capital charges in respect to the underlying pool.
- The transaction could be structured so the lender retains point of sale contact with its borrowers to facilitate its ability to manage borrower relationships. This could be done if the lender retains a role in servicing, perhaps as primary servicer.
- The transaction could also provide a separate income stream to the lender for servicing the transaction.
- This structure could facilitate the ease with dealing with future funding obligations which could be very important for a pool that contains a number of high-touch, heavy transitional (construction?) assets.
- There will be reduced celebrity (and I don’t mean that in a good way). There’s something rather unseemly about selling a pool of loans in an open market transaction. The optics aren’t great. They’re better here.
- The lender could deploy capital into the investment grade securities and put money back to work on a low risk-based capital basis.
- It creates a new path of liquidity for the lender.
- This could habituate the market to the lender’s name as an issuer which could be important for other purposes.
This is a ‘horses for courses’ sort of thing. It’s highly fact and circumstance dependent. Obviously, it’s very sensitive as to whether the transaction will pencil. Note also, there are other alternatives, some substantially more complex than this, including the sale of credit-linked notes (the credit linked note structure is a subject for another day), but this structure has considerable certainty. At a time when liquidity may be dear and multiple modalities to access liquidity are important, this type of thing should receive some considerable attention. And, as I said in my last post, thinking about a novel transaction (or even better doing one) is considerably better than simply waiting for our market to come back.