A curious proposal for securitisation reform has appeared in BIS’s latest quarterly review.
The argument, so it goes, is that securitisation didn’t do so well in the recent financial crisis — structured financial products weren’t immune (or even that well protected) from losses in underlying collateral. Now that resecuritisation seems to be gaining pace, it might be a good idea to try to improve the process — that is, to try to “align the incentives” of those involved in securitisation, as BIS authors Ingo Fender and Janet Mitchell put it.
One way of doing so might be by forcing the originators of securitisations to hold onto a portion of the deal — either the equity tranche, which takes the first losses and is generally considered the riskiest portion — or a sort of vertical cross-slice of all the tranches. Fender and Mitchell explain:
Along these lines, several recent proposals have focused on retention by the originator and/or arranger of some portion of the securitisation. Such a requirement would guarantee that the originator or arranger has some “skin in the game”, providing a direct incentive for prudent behaviour (eg to reliably originate loans based on agreed underwriting standards). The proposal most commonly advanced is to require retention of the equity/first-loss tranche. The idea underlying this requirement is that, by forcing the originator to bear the first losses on the underlying asset pool, the equity tranche will create “highpowered” incentives to exercise due diligence. At the same time, some recent proposals have specified that the originator should hold a share, or vertical “slice”, of the portfolio, perhaps with the idea of balancing the originator’s interests across all tranches with those of the different investor classes.
It should be noted that the idea of tranche retention is not new. In fact, originators in many types of securitisations have traditionally held on to the equity tranche. Over time, however, investors appeared – rightly or wrongly − to become more comfortable with securitised products, leading to a relatively active market in equity tranches. In addition, use of credit derivatives made it possible to at least partially hedge existing equity tranche exposures. As a result, equity tranches, even when originally retained, were increasingly sold or hedged, weakening any incentives that might otherwise have been created for arrangers and originators. While this was known, it was also believed that reputation would play a role in aligning interests, as originators faced the business risk of having investors shy away from their loans if these were deemed to have been originated on the basis of weak underwriting standards.
So much for reputation.
The authors, we would note, don’t seem to be entirely in favour of this method. Instead they’d rather see originators forced to disclose their holdings — rather than be forced to buy part of them.
Along these lines, regulation requiring tranche retention by originators or arrangers is currently under consideration. However, in devising such schemes, care must be taken to appropriately account for trade-offs between market-based and regulatory approaches. In particular, while representing a valuable tool in principle, regulation that imposes a specific retention mechanism is unlikely to adequately align incentives for all transactions. Specifically, retaining equity tranches may not provide strong enough incentives for originators to screen borrowers if downturns are likely and if the retained tranche is thin enough to be exhausted in downturns (ie equity tranche retention is a more effective “fair weather device”). For example, even if originators had expected that housing prices would fall significantly, having them retain the equity tranche of subprime mortgages might not have had the intended effect, unless the equity tranche were very thick. As a result, rigid, “one size fits all” retention requirements that specify both which tranche to retain and how much retention to hold could end up being ineffective or raising costs in ways detrimental to the goal of a sustained market revival.
These observations suggest that forcing originators to disclose the size and nature of any retention may be an alternative to specifying retention amounts. To make such a mechanism work, and irrespective of any formal requirement to actually retain tranches, originators (or arrangers) could be required to disclose the details of any retained exposures, while being granted flexibility regarding tranche width and location in the capital structure. Ideally, such disclosures would then be mandated both at issuance and over the lifetime of any transaction, with a third-party mechanism to validate the information. This, then, would allow markets to flexibly determine the form and size of retention, though with the downside of leaving much of the burden of setting minimum retention amounts with investors.
The European Parliament, US Treasury and IOSCO are nevertheless looking at the idea of forced tranche-holdings, according to BIS.
Thoughts?
Related links:
On killing the market for complex products – FT Alphaville
Curse of the underwriters – FT Alphaville