The combined ESMA / EBA hearing on disclosure, risk retention and homogeneity in the context of the draft RTS published in December was held yesterday (19th Feb). Our overall take-aways are that the extended disclosure obligations and greater standardisation are a clear positive for investors but, on-balance, will likely pose a negative drag on supply-side technicals. In particular, the proposed requirement to disclose credit metrics (PDs/ LGDs) on a loan-level basis will further marginalise public asset-backed funding by the universal banks, in our opinion. We think the risk retention provisions are broadly neutral relative to the status-quo, a number of technical challenges notwithstanding. However, the bigger consideration in this regard is the apparent divergence with rules in the US, where liberalisation of risk retention regulations seems underway.
Some key points:
On disclosure, we see mandatory reporting of loan-level PD/LGDs as a potentially significant challenge, or indeed deterrent, to public securitisations, particularly in the context of corporate assets. This requirement is especially ill-fitting for the leveraged loan CLO market, where we would like to see a specific managed CLO template developed
For corporate asset securitisations in general, we see the name disclosure for non-SMEs (i.e. turnover > EUR 50m) as potentially contravening lender confidentiality obligations and, in the case of leveraged loans, the private nature of transactions
Standardisation of investor reporting is a clear positive in our view, not to mention long overdue
On risk retention, the EBA draft RTS seems balanced in our view, however there are a number of issues that need further clarification, not least around adverse selection given the impact on all securitisations irrespective of retention styles
In terms of who gets access to loan-level disclosures, the lack of definition for “potential investors” (one of the addressee groups for disclosure) potentially creates an unwieldy audience for such loan-level data. We would advocate further clarification in order to limit a proliferation of non-market participants that are able to access such data
A Brief Background
In the framework for the new regulations for securitisations in the EU, the final text of which was published at end December 2017 and is scheduled to be implemented on Jan 1st 2019, ESMA and EBA are mandated to develop regulatory technical standards (RTS) on a number of outstanding issues which are now in the consultation period.
Such RTS will ultimately shape the regulatory framework for asset securitisations in Europe. Its potential effects on the market’s outlook cannot be overstated, therefore. This particular exercise, which included a hearing hosted by the EBA on 19th February, is focussed on writing the rules around portfolio disclosure and asset guidelines, with added refinements to risk retention regulations.
What the RTS says on Disclosure
The disclosure requirements will in principle apply to all securitisations in Europe, enforced via issuer sanctions in the case of any non-compliance. In the following discussion, we focus on public securitisations where the disclosure will be made through regulated securitisation repositories.
As part of this RTS, loan-level reporting obligations are based on the existing ECB templates, which have of course been operational for the last 5 years. However, under this RTS the templates have been updated, with optional fields replaced with additional mandatory fields.
No doubt the provision of additional data fields will be investor-friendly, however we see a number of important reporting challenges for the issuer constituency going forward:
- A new feature under the RTS is extended obligor credit information, specifically around PD/LGD metrics. The requirements are very much styled on bank underwriting practices, and specifically mirrors the internal risk-based (IRB) models of the larger and/or universal financial institutions. We think any requirement forcing such banks to provide and share such forensic credit metrics will only act as a disincentive to securitisation, unless the cost of such funding is comparatively compelling. Moreover, the disclosure of PD/LGD loan-level information, in combination with interest margins, may allow for the reverse engineering of (largely proprietary) underwriting models. Such competitive transparency is unlikely to be welcome among the universal banks using more advanced credit methodologies
- Complying with the new PD/LGD field data will pose challenges for non-bank originators (the one issuer constituency that is driving incremental growth in primary European securitisation volumes) given that they typically capture obligor risks in different ways. Non-banks may be allowed to defer to the “no-data” option but we see this carve-out as being strictly designed for banks using the standardised (as opposed to IRB) approach. More clarity is still needed around eligibility in using this option, but assuming an exemption from PD/LGD data provision is forthcoming for non-banks and standardised banks alike (the latter including mortgage and other ‘monoline’ lenders as well as the challenger banks), we would expect these issuer constituencies to be less impacted by the proposed disclosure rules under the RTS. To be sure, the requirements will be much less onerous and intrusive relative to what would be required from the more advanced IRB banks.
- The disclosure template for corporate asset securitisations includes the requirement to provide obligor names for large corporates. Notably however, based on current EU definitions, ‘large’ corporates are defined as entities with more than 250 employees and revenues above EUR 50m or balance sheets greater than EUR 43m. With such low thresholds, some asset line items in SME loan securitisations may trip into this bucket, requiring originators to disclose their internal ratings and other identifying metrics for these obligors. The Level 1 regulation speaks to the need to maintain confidentiality obligations in article 7.2., but how this reconciles with the new disclosure requirements is as yet uncertain, if not somewhat inexplicable. (We would note that providing such obligor-specific information may contravene client confidentiality laws and/or contractual terms in some jurisdictions). SME securitisations will also generally have obligations to report individual obligor turnover, their NACE industry and NUTS regional codes. While true SME obligor data can in theory be anonymised, the degree of transparency otherwise required would, in our view at least, allow for the potential re-tracing of obligor identities along with lender internal rating opinions
- For leveraged loan CLOs, the draft seeks input if there should be a separate template for ‘managed CLOs’, which we would of course strongly advocate given the fundamental differences between managed CLO securitisations of traded assets and bank balance sheet securitisations of originated, non-traded, corporate lending. All things aside, as CLO issuers are mostly loan managers and not banks, the requirement for PD/LGD disclosure would not be reconcilable with an asset market that operates privately, from borrower credit/ financial profiles to ratings provided by the agencies.
Taken holistically, these requirements as outlined above add further layers of disclosure obligations for asset-backed issuers, which contrasts sharply with statutory reporting norms as well as disclosure rules applicable in any other capital market instrument type. (This line of argument has long been made by the securitisation industry, admittedly). We believe public securitisation activity from the larger, more universal banks will be most impacted, with some use of structured finance technology arguably diverted to private transactions. The impact on smaller banks and non-bank lenders – namely the constituencies that have far fewer funding alternatives other than securitisations – may be less noticeable, with much depending on whether such issuers are able to side-step the obligations to provide more intrusive credit metrics on single obligors. The leveraged loan CLO market, in particular, is very vulnerable to outcomes in this regard, in the absence of any specific templates for the sector.
On a more positive note, we welcome the introduction of standardised investor reporting which should lead to a higher level of comparability. With few exceptions (Dutch prime mortgages and UK credit cards, for instance), many securitisation sectors have been characterised by flawed investor reporting that can often be incomplete and/or otherwise incomparable across instruments given multiple performance definitions and mismatched reporting formats, something that distinguished the European market from its US counterpart right from its very inception. (Transparency in this respect was not helped in some cases by third party data providers publishing slightly different, remodelled data on the same securitised portfolios). The new rules, which extend to demands for pool cut-off dates and investor report publication dates to coincide, is therefore long overdue, in our view. The benefit to market depth and liquidity would be especially noticeable we believe if issuers applied the new reporting formats retrospectively to their legacy programmes.
An arguably minor but still noteworthy point under the RTS is the scope of addressees for which loan-level disclosure will be made accessible. Aside from regulators, the regulation targets ‘investors’ and ‘potential investors’. The definition of ‘investor’ in the regulation is a “natural or legal person holding a securitisation position”, which is of course sufficiently clear. By contrast there is no specific definition of “potential investor” which, in our view, may lead to a far greater audience for securitised loan-level data than anything seen hitherto. To better make our point, we note that the terms of the Level 1 regulation do not, per se, exclude retail investors, notwithstanding the tone elsewhere in the draft RTS (article 9(2)) that seems to limit the scope to institutional rather than individual investors. Moreover, given that its unlikely to be practical for a securitisation repository to validate legitimacy of a “potential investor” or interest on an individual basis, we feel that – in the absence of further definition and/or clarification to this particular point – access to securitisation data may end up being public in all but name.
What the RTS says on Risk Retention
This RTS outlines further refinements to existing risk retention rules in Europe, which apply to all securitised products, historically enforced via regulated buyers (banks, insurance, listed funds etc). The regulation has been amended so that the retention compliance obligation now also extends directly to one of the originator/ sponsor/ original lender to maintain the material net economic interest in any securitisation.
On balance, we do not see the RTS requirements as being disruptive to the status-quo practices in terms of risk retention. Among the noteworthy provisions are further consultations on technical issues such as the (indirect) applicability of retention obligations on European institutions in non-EU jurisdictions, the legal considerations around the financing of retention holdings and definitions related to the ‘sole-purpose’ test where retention is provided by specifically-created vehicles.
In what follows, we limit our discussion to two notable risk retention aspects of the regulation and this RTS – the guidelines around adverse selection and the considerations for the CLO market, in which retention obligations in Europe – as further clarified by the RTS – is diverging from practices in the US market.
Adverse selection is considered along with risk retention given the equivalent principles of alignment of interests. On this point, the Level 1 regulation (article 6(2)) states that any contravention can lead to sanctions, which seems a harsh penalty to us given what is still a grey area. Notwithstanding some helpful clarification (article 16) in the RTS, the concept of “significantly different performance” between securitised and balance sheet assets is still unclear, particularly in the context of low diversified (or lumpy) portfolios where “significant” performance deviations can simply stem from single obligor events rather than outright adverse selection. The RTS contains allowances for non-random asset selection, which we think would be useful in cases such as NPL securitisations or specific regulatory capital transactions, however any such (adverse) cherry-picking will come with the obligation of full disclosure to investors.
The RTS was silent on risk retention for CLO managers. Managed leveraged loan CLOs remain subject to maintaining a material net economic interest of at least 5%, irrespective of the fact that assets are mostly sourced on an arms-length basis from the open market. In itself, this lack of any comment comes as no surprise, however it coincided with a court ruling in the US, in a case brought by the LSTA, which exempts US CLO managers from risk retention requirements. This divergence in regulations is especially noteworthy in the case of CLOs, a sector which has stood out as being global in terms of market technicals as well as structural norms. We think the motivation for US managers to issue dual-compliant CLOs, structured to comply with both US and European risk retention rules, will be significantly tested following the court ruling, particularly considering that there is little evidence, if any, that tapping the European investor base yields much by way of cost benefits via tighter clearing spreads. (We suspect the benefit of deeper liquidity that comes with a global investor base will likely be overlooked). CLO manager economics will surely dominate considerations in this respect, and as such we expect the US CLO market to become more domestically focussed, much like other retention-free asset securitisations (qualifying resi or commercial mortgages, for example). At this stage it is inconceivable to us that European regulators will consider any equally liberal retention regime for European CLO managers in the foreseeable future, and so this divergence in regulations will ultimately impact cross-border CLO market liquidity, in our opinion.
What the RTS says on Homogeneity
The RTS, and subsequent EBA hearing, also covered the issue of asset homogeneity, one of the qualifying requirements for the STS label. In similar vein to a number of other regulatory initiatives, while the suggested RTS is fundamentally sensible and supported by a coherent EBA framework, the definition of homogeneity in itself has been left somewhat wanting. The determination of homogeneity is at any rate far from simple (to be fair, this is more due to Level 1 regulation issues rather than the EBA initiatives), with the requirements for “similar underwriting standards” and “uniform servicing practices” as elements of homogeneity tests further complicating the interpretation in this respect. Further practical clarity would of course be useful, however we doubt that there will be much more by way of definitions provided in the near term.
Disclaimer:
The information in this report is directed only at, and made available only to, persons who are deemed eligible counterparties, and/or professional or qualified institutional investors as defined by financial regulators including the Financial Conduct Authority. The material herein is not intended or suitable for retail clients. The information and opinions contained in this report is to be used solely for informational purposes only, and should not be regarded as an offer, or a solicitation of an offer to buy or sell a security, financial instrument or service discussed herein.
Integer Advisors LLP provides regulated investment advice and arranges or brings about deals in investments and makes arrangements with a view to transactions in investments and as such is authorised by the Financial Conduct Authority (the FCA) to carry out regulated activity under the Financial Services and Markets Act 2000 (FSMA) as set out in in the Financial Services and Markets Act 2000 (Regulated Activities Order) 2001 (RAO). The contents of this report are not intended to be and nor should they be construed as being a financial promotion giving rise to an invitation or inducement to engage in investment activity as provided for in Section 21 of FSMA and the Financial Services and Markets Act 2000 (Financial Promotion) Order 2005 (FPO).
Integer Advisors are not acting as a fiduciary or an adviser and neither we nor any of our data providers or affiliates make any warranties, expressed or implied, as to the accuracy, adequacy, quality or fitness of the information or data for a particular purpose or use. Past performance is not a guide to future performance or returns and no representation or warranty, express or implied, is made regarding future performance or the value of any investments. All recipients of this report agree to never hold Integer Advisors responsible or liable for damages or otherwise arising from any decisions made whatsoever based on information or views available, inferred or expressed in this report.