Can't test our way to trust

Can't test our way to trust

Pic© Felix Burton

Friday 21 November 2014 10:38 London/ 05.38 New York/ 18.38 Tokyo

Mike Manning, co-founder and ceo of DealVector, argues that alignment of interest is the key to RMBS reform

We may be working on the wrong issues. And when you're in a hole, stop digging.

Those were two thoughts I took from SFIG's very well-designed conference on reforming private label RMBS last week. Three moments stood out in particular:
• Treasury's Michael Stegman announcing that the government has decided it is not appropriate to create an "unambiguous fiduciary standard" for trustees.
• Citi noting that it had recently printed a deal with "40 reps and warranties and 85 triggers."
• A participant bemoaning Reg AB II's required disclosure of "only 275 data fields" in new deals and asking: "How do we know that those are the right 275 data fields?"

These are signs of a fundamental breakdown in trust: insisting that someone be labeled as having actual fiduciary obligations; incorporating nearly 100 triggers in a deal; demanding more than 275 data fields. They are more symptom than cure; proof of how thoroughly things have gone off the rails.

Until trust is re-established, no amount of tests or triggers or disclosure will bring the market back. 'Trust, but verify' is viable. 'Distrust and triple-check everything and everyone' is not.

What can rebuild trust?
"Alignment" was the answer that came from several participants. The work that is being done by SFIG and others to develop a better RMBS 3.0 structure is important, but several investors noted that it is impossible to anticipate every possible future problem.

At core, they are asking: "How can I invest if I don't trust you? And how can I trust you if your interests are not aligned with my own?"

Looked at through this lens, current efforts to re-start the market may even be counterproductive. Yes, that 100th trigger or that 276th data field may be the key to alerting investors to problems.

But they may also just add hay to the stack, further obscuring the needle. At the very least, the increased complexity serves as a further barrier to new capital getting comfortable with the asset class.

Consider this: RMBS was ground zero for the 'no skin in the game' problem that was at the heart of the misaligned interests leading to the financial crisis. Yet it has completely escaped the risk retention requirements that have been imposed on other structured asset classes.

Moreover, policymakers have been busy weakening the ultimate 'skin in the game' backstop by lowering the down payment required for qualifying mortgages. No wonder investors are worried.

An interesting exchange developed on the servicer oversight and enforcement roundtable. The servicers described what they were doing to improve servicing and expressed what seemed to be a genuine willingness to disclose more information and to work with investors to assure them that modifications and other decisions were being made in the best interest of bondholders.

However, one noted gently, there does have to be some level of trust. Sometimes oversight degenerates into micromanagement. Any of us who has ever worked for an overbearing boss can sympathise.

"Hold on," responded an investor. "You are asking for a level of trust after we've just been through a period when that trust has been horribly violated." Any of us who has managed an underperforming employee can sympathise with this sentiment as well.

It may be that instead of focusing on governance structures, we should be focusing on incentive structures. This calls into question the whole 'originate to distribute' supply chain that has emerged in RMBS. With each of the functions in a deal - origination, servicing, issuing - dispersed among separate entities, each with its own incentives, RMBS faces unique challenges in assuring investors that their interests will be looked after.

Is this 'distributed' supply chain model viable any longer? Does this explain why RMBS has been so much slower to rebound than other structured categories? Can we maintain the current model, but design different payment schemes to ensure that everyone's interests are actually aligned?

These questions - questions of incentives and rewards, rather than indentures and rules - seem to be the key to bringing back the market. That is because investors, as they made clear at the conference, won't invest their capital again until they are confident that interests are aligned.


×