Open Risk Commentary on Simple Securitisations
Criteria for identifying simple, transparent and comparable securitisations:
(See BIS D304)
Our view is that securitisation is fundamental financial technology and there is no intrinsic technical reason why it could not be harnessed to best serve the functioning of modern economies.
We believe, though, that a comprehensive overhaul of historical securitisation practices is the best means of addressing the stigma that has been attached to it in the follow up to the recent financial crisis. The laudable objective of introducing criteria for simple, transparent and comparable securitisations, (STCS Criteria for short) should aim to achieve a quantum leap in transparency and avoid the risk of perceived ineffectual measures. Technically this would require a visible reduction of model risk for prospective investors, expressed for example in tangible new ability to perform relevant risk analyses.
We would suggest to use the opportunity of defining the STCS criteria to help create genuinely new originate-to-distribute (primary bond issuance platforms) where sophisticated investors have access and make use of all material risk information. These platforms should make maximal use of powerful and cost-effective new IT technologies and should not be unduly hampered by the limited systems and correspondingly adapted practices of past decades.
In order to put our detailed commentary on the criteria in further context we would like to add the following specifications on what we believe should be desirable economic characteristics of the STCS compliant instruments:
- STCS bonds should be diversified credit instruments reflecting as accurately as possible credit performance, in the bulk, of key credit asset classes that are not otherwise accessible to investors. This has implications for the nature of eligible assets, their portfolio profile and the possible co-mingling of other risks in the credit risk assessment.
- STCS structures should create price transparency and investment opportunities in both the total credit pool being securitised (pass-through possibility) and the prioritized repayment (subordinated / tranche) exposures, not just the latter.
- STCS structures should by construction not allow arbitrage between total pool investments and sum of tranche investments. In this manner, credit enhancement via subordination enables price discovery around expected credit volatility which is a fundamental economic variable.
Detailed points which we suggest should augment the current criteria
A. Asset Risk / 1. Nature of the assets
Assets being securitised should - as a matter of principle - not include other securities, irrespective of whether those are simple or structured credit exposures.
STC eligible assets should have strictly limited prepayment and/or drawdown risks. The presence of embedded optionality significantly complicates fundamental credit risk analysis (requires competing risks frameworks) and turns securitisation bonds into dual-asset (credit + interest rate risk) fixed income products.
Assets should not form unusual or hidden portfolio risks due to large concentrations in few credits or disparate maturity profiles leading to refinancing risks. Containing the magnitude of such risks could be achieved by means of standardized concentration risk and amortization criteria
A. Asset Risk / 2. Asset performance history
Model validation: There should be qualitative and quantitative criteria on the availability of historical data that are commensurate e.g., with those applicable, to internal credit model development for IRB banks.
A. Asset Risk / 4. Consistency of underwriting
Documentation of underwriting standards: Relevant “underwriting standards” applicable to the underlying credit instruments should be disclosed to investors (e.g,. in the form of databases with key loan / client characteristics)
A. Asset Risk / 6. Initial and ongoing data
Definition of a golden standard for data requirement: The golden standard for enabling best practice initial and ongoing due diligence is the provision of sufficient performance data (Point A.2) and client / loan characteristics (Point A.4) to enable investors (with the possible use of additional external datasets) to construct their own credit risk models for individual credits and/or granular credit pools.
B. Structural risk / 9. Payment priorities and observability
Limited number of liabilities: Liability waterfalls should be drastically simplified, with only a few (3-4) major liability classes that span an approximately logarithmic risk reduction schedule. Large number of liabilities (issued bonds) with elaborate prioritisation simply ensures that the risk profile of each of the bonds is extremely sensitive to model risk and thus proliferates meaningless credit ratings.
Pass-through transparency: Liabilities should always include a pass-through component that would ensure there is a complete market view on the credit risk profile of the securitised pools.
B. Structural risk / 11. Documentation disclosure and legal review
Transaction templates: Each STCS bond issuance should adhere to one of a set of well defined STCS documentation templates (standardized term sheets) with variations within STCS class limited to certain quantitative parameters. The use of such templates would dramatically reduce the costs of analysis and improve comparability across time and portfolios.
Documentation Language: All material terms defining the risk profile of the bonds should be described concisely in readable prose, in contiguous pages and with clear supporting flowcharts. The terms should also be described using mathematical formulae.
Cashflow Model Validation: The technical implementation (encoding) of cashflow models should be easy to validate against documented mathematical expressions.