The IDSA believes the suggested RTS provides a structured framework to help determine homogeneity. We believe that the EBA is correct to state that “given the broad scope of the asset categories, belonging to one such asset category does not render the underlying exposures sufficiently homogeneous. Additional criteria, largely in line with the current market practice, should therefore be applied, in the form of risk factors, the application of which would result in further differentiation of exposures within the respective asset category.”
From a portfolio concentration and diversity perspective, we have a concern with the main approach chosen to determine the homogeneity of the collateral pool and believe that there are strong arguments to proceed with the hybrid version. We agree that the use of similar underwriting standards, methods and criterion in conjunction with uniform servicing procedures should provide the base for determining homogeneity. Combining this with a non-exhaustive list of asset categories ensures a further degree of common and shared risk profiles. The main approach also includes using a pre-defined list of risk factors to further classify the underlying assets. While this is likely to produce an extremely homogeneous pool of assets, it also has the danger of producing a very concentrated pool of homogeneous assets which will reduce the level of diversity within the pool and could well increase the overall riskiness of the pool as it becomes too concentrated in a similar type of asset.
The IDSA believes that risk factor (d) could potentially cause a degree of undue concentration risk and reduce diversity within the pool. The IDSA believes that it is worth considering an alternative option to determine the homogeneity in the pool. Under an alternative proposal, “the homogeneity would be defined by means of compliance with a set of detailed criteria, capturing a wide spectrum of potential sources of heterogeneity of the underlying exposures linked to their various cash flow, contractual, credit risk, prepayment and other characteristics (including, for example, with respect to the currency, maturity, or minimum credit quality).”
The EBA states that this approach would be inconsistent with the treatment of other classes such as covered bonds where the assessment of the assets is performed at the level of the asset category (i.e. it is a common market practice that the cover pool assets are composed of one primary asset class, such as, for example, residential loans). However, this does not seem to take account of the fact that under these proposals there is the additional requirement to take account of “similar underwriting standards” and “uniform servicing practices” as elements of the homogeneity test.
The RTS states that homogeneity for the purposes of STS securitisation should not provide incentives that would prevent the originator from structuring a diversified portfolio, nor should it lead to excessive concentration in the portfolios, for example to exposures to obligors in a specific geographical area, or to a specific type of obligors. We believe that the use of too many criterion and categories may result in a less diverse portfolio.
The pool of underlying exposures should only contain exposures of one asset category which share similar characteristics with respect to the type of obligor, the credit facility, the collateral, the repayment characteristics or other factors, because such similarities enable the investor to assess the pool of underlying exposures on the basis of common methodologies and parameters.
The RTS lists among its risk factors repayment/amortisation, industrial sector, jurisdiction and, governing law. The draft standards outline an example of a homogeneous pool that has been differentiated based on relevant risk factors. In this example, it states that “The pool of underlying exposures would thus only contain exposures of non-income producing residential mortgages secured by residential property located in one specific jurisdiction (and could not be mixed with exposures of income-producing loans secured by residential property in that of other jurisdictions).” It states that this should result in a pool of exposures that have similar risk profiles and cash flow characteristics, enabling the investor to assess the underlying risks on the basis of common methodologies and parameters.
While the IDSA can see the merit in this proposal we believe that the use of the risk factors may have three unintended negative consequences:
1. Concentration risks/less diverse
2. Pools not reflective of the business models of the originators
3. Challenges in drafting the key risk sections in the prospectus
By applying so many restrictions through the risk factors, the collateral pools will, by design, be less diverse and more concentrated which may actually contribute to a riskier pool than a more diverse one. The pool will be subject not only to similar asset types but also similar lending products in the one jurisdiction. Different lending products will have different characteristics, which may well have different performances at different stages of the economic cycle. Diversity is a positive attribute and in general, leads to lower levels of risk. By overly restricting the degree of diversity, there is a real concern that more concentrated portfolios may be subject to event risk and have a more volatile performance than more diversified portfolios.
Limiting the collateral pools by the proposed risk factors may, in many instances, result in a collateral pool that is not reflective of the actual business transacted by the originator.
Many originators originated many different products under one set of credit and risk management policies and underwriting standards. Many originators, due to their limited scale may not have a sufficiently large homogeneous portfolio, as defined by the risk characteristics, to achieve the economies of scale to facilitate a cost-effective securitisation and thus preclude them from participating in the market. Indeed, as the Capital Markets Union in Europe develops further, it is expected that there will be significantly more cross-border lending in the real economy. As originators expand their footprint outside their home market they too will be constrained as they will not be able to issue cross-border STS securitisations and may also suffer from a scale problem.
The IDSA also has a concern with the impact of using risk factors in determining the homogeneity of a pool and then having to include such risk factors in a prospectus where the number of key risks that can be discussed are limited. The ranking of the risk factors, which is already considered challenging, is likely to become more difficult.
The concept of allowing adverse selection in the transfer of assets to the SSPE outlined in the draft Regulatory Technical Standards specifying the requirements for originators, sponsors and original lenders relating to risk retention allows adverse selection as long as this is “is clearly and conspicuously communicated in writing to the competent authorities, investors and potential investors prior to the investment being made.” Adopting similar language in relation to disclosure of these risks would also allow flexibility in selecting assets to be transferred to the SSPE. The full disclosure would help transparency and ensure that the risks are highlighted.
Overall, we believe that the adoption of a hybrid proposal would contribute to a better functioning market
See 3 Above.
ESMA in its Consultation on its Draft technical standards on disclosure requirements, operational standards, and access conditions under the Securitisation Regulation has considered the possibility of developing additional underlying exposure templates for other categories of underlying exposures found in non-ABCP securitisations.
Based on this analysis, ESMA identified three possible categories: CDOs, Whole Business Securitisations and Rare (across the EU) and idiosyncratic underlying exposure types: this category includes healthcare receivables, funds recovered from electricity tariff deficits, student loan payments, dealer floorplan receivables, and any other underlying exposure type not covered in the above-mentioned categories and templates.
Following a mandate received from the European Commission, the EBA has also developed a set of draft standardised reporting templates for NPL exposures.
It would appear logical that the asset categories used by the EBA and the templates developed by ESMA should be comparable.