Prudential issues in securitisation

Charles Littrell, Executive General Manager - Australian Securitisation Forum, Sydney

APRA’s statutory mission is to balance regulated entity safety and Australian systemic stability with financial system competitiveness, efficiency, contestability, and competitive neutrality. In this work, we tend to focus first upon safety and stability, and regard the other statutory balancing items as constraints. That is, APRA actively promotes safety and stability, but attempts to avoid damaging efficiency and competitiveness.

Securitisation is an interesting prudential proposition for APRA. Up until about 2004, this technique generated clear benefits for competitiveness, efficiency, contestability, and neutrality. The first two points are demonstrated by the fall in home lending margins from the early 1990s to the mid 2000s. The second two benefits are demonstrated by the ease with which relatively small regulated entities, and entities unregulated by APRA, were able to enter the home lending market and compete effectively against the larger banks.

As for safety and systemic stability, in its earlier and simpler incarnation, securitisation was not a problem. The ADIs participating in this market generally maintained good lending discipline, abetted by the lenders mortgage insurance companies supporting the market. In Australia at least, unregulated lenders generally exercised some common sense in their lending and loan management.

We saw in America and Europe, however, that securitisation in the 2000s increasingly became a vehicle which supported over-complexity, reckless and too often fraudulent lending, and gross conflict of interest. The 2007/08 phase of the global financial crisis was not caused by securitisation, but securitisation structures and re-securitisation structures featured prominently in the capital losses and illiquidity associated with that crisis.

These issues affected the Australian financial markets, but mainly around the edges. Some ADIs, insurance companies, and super funds lost money on CDO-squareds and the like, often backed by various tranches of securitisations. Probably more importantly, the collapse of investor demand for securitised paper meant that the smaller and mid-sized ADIs lost some of their competitiveness against the larger banks in home lending.

Having lived through this experience, APRA remains of the view that securitisation is more useful than it is dangerous. But we have become considerably less tolerant of the over-complication that crept into this financial product. We also observe that some ADIs have developed the habit of following the letter, but not the spirit, of the prudential requirements applicable to securitisation.

Many of you will recall that as part of the 2008 roll-out of Basel II, APRA re-issued APS 120, our prudential standard on securitisation. APRA elected to move from pre-vetting all securitisation transactions, to reviewing transactions in hindsight as part of our ongoing supervision. The idea was that ADI executives and boards should self-assess compliance against the relevant standards, and apply the relevant capital and other requirements.

We had planned to review industry compliance with APS 120 during 2008, but other priorities intruded. It was not until late 2009 and early 2010 that APRA undertook a more systematic look at industry compliance with APS 120. What we found was often not pleasing. Among other things:

1. Some ADIs were progressing securitisations and claiming capital relief, without meeting the basic documentation requirements associated with self-assessment;
2. Nearly all ADIs were assessing compliance as a legal matter, relying upon legal opinions as opposed to their own judgement as to matters such as effective risk transfer;
3. Faced with an effective shutdown of risk appetite by investors, some ADIs retained all the junior tranche in their securitisations, and therefore did not transfer much if any credit risk. Many of these ADIs nonetheless claimed to have taken the relevant risk assets and capital requirement off balance sheet, even though this is clearly an incorrect treatment. Other ADIs were adopting the more sensible approach of retaining junior tranches, but not claiming any capital relief; and finally
4. Around the edges of the ADI industry, we were seeing signs that some issuers were placing junior tranches in non-commercial fashion, for example with corporate cousins or by arbitraging with other ADIs.

APRA is a principles-based regulator, and one problem with securitisation is that it is essentially an artificial, heavily rules-based concept. This has led to the practice by some in the market of "what the rules allow", rather than "what is economically sensible". APRA regards the basic concessions in securitisation, the ability to remove all capital requirements on the underlying assets, and the permission to pledge assets, as extraordinary benefits. We do not propose to award these benefits to structures and issuers that do not merit them.

By late 2010, it had become apparent that placing the equity tranche or so-called "B notes" had become a key issue in securitisation. APRA issued some interim arrangements in December 2010 to allow ADIs to continue securitising, even if they couldn’t sell their B notes. The basic prudential treatment was:

1. That ADIs not seeking capital relief but only funding could simply hold on to the B notes, and not claim a capital benefit.
2. ADIs seeking capital relief but holding B notes could deduct from Tier I the value of the B notes held.

During 2011 APRA has spent considerable energy developing its Basel 2.5 and Basel III proposals. Along the way, we have considered how best to approach APS 120. Competing priorities mean that an APS 120 review is some way away.

Instead, we recently issued an update to the 2010 interim securitisation arrangements. The main issues were:

1. We have clarified that ADIs can pursue funding-only securitisations; and
2. We propose to install an anti-arbitrage rule that prevents ADIs from trading B notes and making the capital deduction disappear.

During 2012, we hope to find the necessary time to commence a complete review of APS 120. It would be fair to say that the current strategy, which is to chase a complicated product with complicated rules, has not worked as well as we might have liked. Accordingly, any reformed APS 120 is likely to feature a simpler approach, but with more supervisory flexibility.

I offer the following personal thoughts not as a preview of any new APS 120, but to give a sense of the issues with which we are wrestling.

First, we clearly will need to coordinate APS 120 with the world’s moves towards skin in the game requirements.

Second, we need to regularise the funding-only approach to securitisation. Should we, for example, mandate a complying structure, or allow the industry to continue adding bells and whistles?

Third, it is not at all obvious from APRA’s systemic stability viewpoint, and from our superannuation and insurance regulator viewpoint, that it makes sense for ADIs to place junior tranches with less informed investors. We need to think harder about the cross industry treatment of such paper.

In a similar vein, a securitisation structure must have at least two tranches to be a securitisation. Is there any good prudential reason to have more than two tranches? Who could hold the mezzanine tranches, with a knowledge or funding advantage over the originating ADIs? One can see a reasonably clear role for tranches differing in maturity and payoff profile, but the case for several levels of credit risk is much less clear.

On the other hand, it seems clear that most, and possibly all, resecuritisations fail the safety vs. efficiency and competition test. Why do these arrangements exist, other than to sell apparently safe but actually risky investments to under-informed or potentially deceived investors?

Moving to the other side of the balance sheet, Basel III prescribes a complicated set of risk weights to ADIs holding securitised paper. APRA will take an approach that complies with Basel III, but we are likely to consider less complicated applications. We do not favour ADIs buying junior securitisation tranches from other ADIs, for example, and our capital treatments reflect this view.

In summary, APRA expects that the securitisation market will continue to recover, and that funding-only securitisations will take more of a role. It is not yet clear how securitisations for capital relief will develop. APRA’s intent is to provide a prudential framework in which such structures can work, but in which artificial imitations of true risk transfer structures do not work.

I take this opportunity to comment briefly on covered bonds. In recent weeks the Government has legislated to allow ADIs to issue covered bonds, up to a limit of 8 per cent of each ADI’s Australian assets. APRA immediately removed the covered bond prohibition from APS 120, and recently we issued a short discussion paper and draft APS 121 to address covered bond issues. In reviewing the initial deal structures, we see a few issues that will benefit from prudential guidance.

1. We need to ensure, typically through a right of first refusal granted to the originating ADI, that the covered bond trustee does not dispose of assets at fire sale prices. When fully up and running, an ADI could have close to half its equity exposed to the overcollateralisation assets in the covered bond trust. In a default we want this equity to come back to the originating ADI, not to be dissipated in a fire sale.

2. We also view with some disquiet the proposed practice of placing extra assets, above the cover pool, in the covered bond security vehicle, and not counting against the 8 per cent limit. We have had representations from several ADIs that these assets can come back at any time. At this stage, however, we are unconvinced that assets placed in the potential control of a security trustee in another vehicle are just as depositor-friendly as assets held directly within an ADI.

3. We have clarified that assets pledged above the 8 per cent limit will attract an equity deduction.

As a final point, I note that between securitisation, covered bonds, more collateral for trading exposures, including the need to collateralise central counterparty exposures, and the probable exploration of repos in the context of ADI liquidity and other needs, we may be observing an historic shift from banks mainly as unsecured borrowers, to banks pledging a great deal of collateral. Although each of these initiatives individually may give an ADI cheaper funding or better trading terms, a whole industry with lots of collateral pledged is most unlikely to make the remaining depositors and unsecured creditors safer. This is an issue that APRA and other regulators will need to wrestle with over the next several years.

Thank you for your attention.