APRA update: Reforming the prudential framework for securitisation

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

It is good to see you all again, as we once again discuss prudential reform in securitisation. Since speaking on the same topic last year, APRA has released a discussion paper, and received considerable feedback from industry and interested stakeholders. I note that the ASF has been central to coordinating this feedback, which has assisted APRA in refining its approach.

At last year’s conference, I outlined APRA’s proposed principles for securitisation reform:

• the securitisation prudential regime explicitly allows for both funding-only and capital relief structures;
• credit risks are both clearly assigned and, for APRA-regulated entities, properly capitalised;
• securitisation reduces rather than increases systemic risk;
• the maturity profile of securitisations must appropriately match the maturity profile of the underlying assets or, for short-term assets, match the maturity profile of winding-down or divesting the relevant line of business; and
• securitisation is simple, transparent and low risk.

Our work over the past year has confirmed APRA’s view that these are appropriate principles for Australian securitisation. Furthermore, our sense is that the securitisation industry is in the main comfortable with these principles, although with many suggestions on the details.

Some service providers to the securitisation industry have argued that APRA should cater for more complexity, but issuers have broadly argued that simplicity is fine, and we should get on with the reforms so they can issue a lot more paper. We tend to side with issuers on this point. APRA’s thinking on securitisation has always been that we view this tool as a potentially useful way to increase efficiency and competition in Australian financial markets, but we are unattracted to the complexity and opacity that too often has attached to past securitisation structures.

APRA’s intention with securitisation is threefold:

• first, we want to facilitate a much larger, but very simple and safe, funding-only market;
• second, we want to also facilitate a capital-relief securitisation market; but
• in both cases, we want to impose a simpler and safer prudential framework than has evolved internationally.

APRA’s sense is that the best prudential strategy for Australian securitisation is to provide a framework that adopts the simplest possible workable structure. That is, given that securitisation necessarily needs some complexity and perhaps some regulatory flexibility, we propose to provide enough to allow a large market to grow, but without the excesses that existed pre-crisis.

Today, I propose to outline APRA’s approach to several issues that have arisen in our consultation. These fall broadly into four categories:

• structural questions, mainly about master trusts and warehouses;
• significance questions, mainly for the larger issuers;
• skin-in-the-game and subordinated tranche issues; and
• timing and transition.

In its discussion paper, APRA indicated that we would consider master trusts for funding-only securitisations. The response from industry has unsurprisingly urged us to allow originators to optimise master trusts on behalf of investors, as this maximises the potential issuance volume and reduces the issuer cost.

Structural issues

The only drawbacks to such flexibility, sadly, are that unduly flexible master trust arrangements may increase prudential risks for ADIs, and degrade the protection available for ADI depositors.

Balancing APRA and industry desire to promote master trusts as a basis for a much larger securitisation market with APRA’s desire not to increase risks for ADIs and depositors is probably the trickiest policy question remaining in front of us.

At this point last year we had two large issues to resolve for master trusts: the treatment of seller interests in the securitisation vehicle, and the rules regarding date-based calls and bullet maturities. I am perhaps tempting fate to assert that we are evolving a reasonable balance on the first issue. The evolved balance is that seller-contributed or retained assets in the vehicle must rank at least pari passu with the most senior investor class in the vehicle. This ranking, however, would apply to priority in a wind-up or similar circumstances. In the revolving phase, APRA is considering allowing the seller repayment to flow after other creditors, and in a less predictable fashion. This approach should substantially facilitate master trust structuring—but we have yet to fully satisfy ourselves that this facilitation is sufficiently safe, so more work remains.

The bullet- or date-based call issue is more difficult. The main difficulty is that the contractual structure does not match market reality for bullets, even of the allegedly “soft” variety. That is: if APRA allows master trusts to issue third party paper with a heavily signalled maturity date, sufficient to allow such paper into the relevant bond indices, then there is a firm market expectation that the instruments will in fact be paid on the signalled due date. This expectation holds even if the contract allows for an extension of maturity due to unexpected shortfalls in the vehicle’s cash flow.

The problem here is quite similar to the 1998 to 2008 vintage of preference equity and subordinated debt issued by many banks. These capital instruments were contractually long term or perpetual but generally had a strongly signalled maturity date. Failing to meet the market’s expectations, APRA was frequently told, would have sent an adverse signal about the health of the institution.

There is also the consideration that APRA intends to integrate securitisation cash flows into its now-reformed liquidity requirements for ADIs. This integration goes some way towards addressing the bullet maturity risk, but not necessarily in a fashion that satisfies the idea that securitisation structures can stand on their own, without ad hoc support from the originator.

APRA has not yet determined its final position on date-based calls. We remain hopeful that we can find a balance that grows the market, without threatening ADI or depositor safety. In any event, we have asked the ASF and some other parties to suggest structures and term sheets that would address APRA’s concerns.

The main issue for APRA in warehouses is that current arrangements may allow a small ADI to sell loans to a large ADI, no third party securities are issued, and neither ADI holds substantial capital against the loans. This is clearly an imprudent outcome, and APRA considers there are sound grounds to remove this arrangement from the prudential framework. It is not yet quite clear how we might replace the current arrangements.

One candidate would be to treat the acquiring ADI as the loan holder, with capital requirement to match. If the acquiring ADI then undertakes a securitisation, capital and liquidity benefits will flow as if the warehousing ADI was the originator.

Another candidate is to proceed with our proposed 12 month hold test. We are also considering whether warehouses should be funding-only structures, as the major prudential impediments seem to flow from capital treatment.

I note that a number of industry participants have asserted that it is necessary for originating ADIs to be able to place new loans directly into the warehouse, rather than simply build up a loan portfolio on the ADI balance sheet, against the day when a critical mass is reached. I must say that APRA is unsympathetic to this argument. Any ADI lacking sufficient capital and liquidity to support new lending at the margin, does not need relief through lax securitisation rules. Rather, that ADI needs more capital, more funding, or perhaps a larger and stronger merger partner.

APRA does not propose to create a securitisation warehouse regime that allows the industry to retain the loans but artificially lose the capital requirement. We need to figure out how to link any capital benefits to true risk transfer to a non-ADI third party, or alternatively accept that much of what now happens in a warehouse is really just secured funding from a larger ADI to a smaller ADI. As another alternative, we could consider that warehousing is simply a loan sale from a smaller to a larger ADI. We consider that APRA already has satisfactory capital arrangements outside securitisation that covers such transactions.

Significance questions

Since 2009, APRA has increasingly considered securitisation a liquidity management vehicle, as well as a credit risk transfer arrangement. To the extent that an ADI can place long-term funding with suitable investor groups, then a reformed securitisation market has the potential to appreciably improve the Australian banking system’s liquidity.

During 2014 APRA has concluded its work on the first phase of Basel III liquidity reforms, covering the Liquidity Coverage Ratio (LCR), and the Reserve Bank of Australia’s (RBA’s) provision of a committed liquidity facility (CLF). Access to the CLF requires an ADI to demonstrate to APRA that it has taken all reasonable steps to minimise reliance upon public sector support, in favour of finding liquidity in private markets.

Now clearly, if APRA facilitates a potentially much larger securitisation market, then we will expect this potential to convert to reality. In that vein we have commenced discussions with some of the larger ADIs, to determine their intentions regarding expanding their securitisation-sourced term funding as a means of meeting the ‘all reasonable steps’ test.

Is APRA more conservative?

APRA’s statutory mission includes the legislative instructions to balance safety with efficiency and competition, and to promote financial system stability in Australia. For many years, APRA has taken the view that the appropriate balance and optimal stability will be found some distance from the international lowest common denominator. This is because we consider that Australia is best served by a materially lower bank and insurance failure rate than is the international average. Accordingly, APRA routinely applies somewhat more conservative definitions and higher requirements than may apply internationally to relevant calculations for capital, liquidity, loss reserving, and the like. Many peer regulators do likewise.

We also and at least as importantly take a more proactive and when necessary interventionist approach to supervision. This does not mean that Australia will never suffer a prudential failure. Such failures are statistically inevitable, whatever APRA does. Our approach should mean, however, that such failures are relatively rare and less likely to generate adverse systemic shocks, than would otherwise be the case.

APRA’s approach to securitisation will be consistent with APRA’s general approach. Today I’ll touch briefly on several relevant areas where this is likely to be the case.

Let’s start with credit risk retention or ‘skin-in-the-game’. APRA’s proposed approach is simple:

• for funding-only transactions, it would be appropriate for the originator to hold all the subordinated instruments and all the capital requirements;
• for capital relief transactions, the originator should never reduce below 20 per cent risk and capital for skin-in-the-game; and
• the regulatory capital relief should tie to the least-sold subordinated tranche.

This approach allows ADIs to achieve quite a lot of capital relief, but also ensures that ADI lenders will always retain material credit risk. We haven’t yet been given any persuasive arguments on why this approach is not the best mix of simplicity and prudence. It is certain that the U.S. and European skin-in-the-game rules will differ from APRA’s proposed approach, and are likely less conservative, but this is not an argument for APRA to adopt weaker rules on skin-in-the-game.

Similarly, APRA has proposed that the capital treatment for ADIs holding anything other than the most senior tranche in a securitisation is very simple: 100 per cent Common Equity Tier 1 (CET1) deduction. Many respondents have pointed out that this is not the approach proposed by the Basel Committee, or taken in offshore jurisdictions. We agree that APRA is following a simpler approach, but again we have yet to hear a persuasive argument for why ADIs should be anything other than firmly discouraged from investing in junior securitised paper.

As I mentioned earlier, APRA is inclined towards a reformed securitisation framework that should set the simplest workable rule, and resist any move towards additional complexity. In this context simple usually equates to conservative, and this is likely to best serve the diverse interests of the banking sector, the securitisation industry and, importantly, end-investors.

This is the third speech I have given at an ASF meeting on this round of securitisation reform. I expect that at least one and probably two more annual speeches will be necessary before we complete. My tentative title for the 2015 edition is: “The reforms are starting shortly, go grow safely”. The 2016 edition would be: “The reforms are in place, and my, how big and strong the market has become.”

Timing and transition

We are proceeding somewhat more slowly than usual on securitisation reform, for several reasons. First and most pragmatically, there is rather a lot on a bank regulator’s plate at the moment. The Basel III reforms are largely in place, but several more initiatives are flowing from Basel in the near future, including a global securitisation framework. There is also the local matter of the Financial System Inquiry, and APRA would like to see what the Inquiry has to say, if anything, about securitisation.

There is also the consideration that securitisation is a somewhat complicated proposition, that APRA is not simply following a well-established global standard, and it behoves both APRA and industry to take our time and get this reform right.

I would hope that APRA will complete a new APS 120 during calendar 2015, with some transitional provisions. I expect that we will propose APRA’s typical approach to transition, which can be summarised as:

• new transactions comply with the reformed rules from issue date;
• older transactions are either run off, closed out, or restructured as quickly as is consistent with reasonable costs; and
• no permanent grandfathering applies.

If past experience holds, the transition of legacy transactions will require two to three years from the effective date of the new APS 120. I note that many extant transactions may in fact comply with the reformed prudential standard, particularly if they are funding-only amortising structures.

Summary

Given time constraints I have skipped several technical details today, so we won’t have a rousing discussion of, for example, trust-back arrangements. APRA is working on these details in conjunction with industry.

My best estimate at this point is that APRA will produce a reformed APS 120 for consultation during 2015, and the new prudential standard will both facilitate a larger securitisation market, and protect ADIs and depositors from some of the perils that may arise from this market. Time, and not all that much time, will tell.

In closing I emphasise that APRA is working closely with the securitisation industry and its stakeholders in refining our reform proposals, and we expect to continue with this approach.

Thank you for your attention, and I will now take your questions.