Response to consultation on draft RTS on the determination by originator institutions of the exposure value of SES in securitisations
Go back
However, we note that the approach designed in the draft of the RTS does not seem to prevent regulatory arbitrage, but it includes elements that might even promote it. In fact, prudent levels of SES are “punished” by high exposure values and subsequently weaker capital relief, while high levels of SES do offer relative lower exposure values for excess spread and subsequently achieve better capital relief.
In our view, this would represent an incentive for regulatory arbitrage than a restriction. This is due to the fact that the exposure value for future excess spread is limited by the total expected losses according to the asset model approach instead of being based on the amount exceeding expected losses. Increasing excess spread levels would provide substantial credit support to the tranches and therefore strongly improve the risk profile of the tranches. To illustrate this, we would like to refer to the representative transaction on page 25 of the draft RTS. The following table (see attached document) shows that the exposure value does only increase marginally with higher excess spread levels:
As mentioned, this would promote regulatory arbitrage rather than restricting it.
In addition, this would hamper the level playing field between traditional and synthetic transactions, as only synthetic transactions would be affected, while an equivalent traditional transaction would not.
Having in mind the stated aim of both the CRR Quick Fix and the draft RTS, we believe that an approach along the following lines would be more appropriate, economically meaningful and also consistent with the Level-1 text.
By way of derogation from the treatment of traditional securitizations under the CRR one can argue that future excess spread of a securitization covering the expected loss is already risk-weighted under the existing rules of Chapter 5. As indicated in 8 (ii) of the draft RTS in a traditional securitization the originator is giving up future income of securitized assets. This does include the excess spread portion of the future income of assets, which is available to cover losses of the securitized portfolio within the traditional securitization structure in the first place. This credit support in form of future excess spread is of course affecting the risk profile of the tranches and is also considered in pricing.
Subsequently applying the rules of Chapter 5 for the calculation of the risk-weights of tranches implies that credit support provided by future excess spread is already considered in the risk weighting of tranches for traditional securitizations. This is supported by the fact that SEC-SA and SEC-IRBA effectively assign a multiple of the risk-weighted assets of the underlying portfolio to the various tranches according to an exponential function, therefore already considering the expected loss of the underlying portfolio.
So, while in a traditional securitization future income (including the excess spread portion of it) is given up by the originator, in a synthetic securitization the future income (including the excess spread portion of it) stays with the originator, as it is correctly stated in para. 8 (ii) page 6, if no synthetic excess spread is provided. A synthetic securitization therefore is equivalent to a traditional securitization if and only if the originator assigns the excess spread portion of the future income to the structure to cover expected losses in the portfolio in form of a synthetic excess spread. This is both true for the originator’s position in terms of future income as well as in terms of the risk profile of the tranches. In this context we would like to highlight the last paragraph of Art. 248 (e). In our view by way of derogation from the treatment of traditional securitizations it is economically intuitive and well arguable that future synthetic excess spread up to (!) the expected loss is already risk-weighted under the rules of Chapter 5, while any synthetic excess spread portion exceeding the expected loss is not.
We therefore would consider it to be economically meaningful and in line both with the Level-1 text and the intentions of the legislators to focus on future excess spread exceeding the expected loss of the securitized portfolio (plus past/current excess spread which is already considered in the income statement/balance sheet) for calculating the exposure value instead of the future excess spread up to the expected loss. In our view such an interpretation would also better serve the goal to prevent regulatory arbitrage by requiring risk-weighing of excessive future synthetic excess spread.
Finally, the current draft has an excessive undue punitive effect on smaller banks by requiring substantial risk-weighting for prudent levels of synthetic excess spread: due to their smaller loan books both in size and number of positions, securitized portfolios are often longer in average life and therefore more dependent on excess, while a prudent level of excess spread is economically justified. Also securitized portfolios of smaller banks are often more focused towards SMEs, which could reveal a riskier profile, generating higher income, having higher expected losses and again as a result being more dependent on a prudent level of synthetic excess spread.
In case the approach currently taken in the draft RTS is nevertheless maintained, we would also like to note the following.
There is an evident lack of a grandfathering or phase-in provision for existing transactions. Existing transactions structured with excess spread will become uneconomic and even lose their complete RWA relief in the worst case (especially EIF structures include excess spread and a retained first loss tranche that have the risk of falling below 1250% RW with future excess spread inclusion in SEC IRBA formula and thus not be capital deducted anymore but with high RWAs to be considered).
This could trigger a high number of regulatory calls. The consequence will be a big negative effect from loss of capital relief for many European banks in a difficult environment at effectiveness of the proposed paper if the future excess spread calculation is not reconsidered.
This would again be especially burdensome for smaller banks as due to typically smaller transaction sizes and often longer WALs of securitized portfolios sunk costs due to regulatory calls would be higher from a relative perspective.
Q1. Do respondents find the provisions clear enough or would any additional clarification be needed on any aspect?
NAQ2. Do you agree with the possibility of choosing between the full and the simplified model approaches in a consistent manner?
NAQ3. Instead, would you favour that the RTS consider only one method (i.e. the full model approach or the simplified model approach) for the calculation of the exposure value of the synthetic excess spread of the future periods?
NAQ4. Do you agree with the specifications of the asset model made in Article 3?
NAQ5. Do you agree with the specifications for the determination of the relevant losses made in Article 5?
NAQ6. Do you agree with the calculation of the exposure value of synthetic excess spread for future periods made in Article 6?
NAQ7. Shall the average of the scenarios be made in a different way for UIOLI and trapped mechanisms (e.g. back-loaded and evenly-loaded only for UIOLI mechanisms, and front-loaded and evenly-loaded for trapped mechanisms)?
NAQ8. Do you agree with the specification of the simplified model approach made in Article 7?
NAQ9. Do you consider that the formula can be further simplified (e.g. by using the maturity of the credit protection multiplied by a conservative scalar instead of WAL)?
NAQ10. Do you agree with the scalar assigned for UIOLI mechanisms? If not, please provide empirical evidence that justifies a different scalar based on the different loss absorbing capacity of UIOLI vs trapped mechanisms.
NAWould you favour that approach? If so, how do you think that this rolling-window approach for calculating UIOLI SES will affect the efficiency and viability of synthetic transactions in comparison with the current supervisory practices? Please justify your response with specific illustrative examples or data.
NAQ12. Do you agree with the treatment of the ex-post SES of future periods in the RTS? If not, please provide rationale and data supporting your views
NAQ13.Do you have any other comments on these draft RTS?
As indicated in the consultation paper, Synthetic Excess Spread (SES) is considered a securitisation position subject to capital requirements under the CRR, because of concerns regarding the regulatory arbitrage that SES may imply.However, we note that the approach designed in the draft of the RTS does not seem to prevent regulatory arbitrage, but it includes elements that might even promote it. In fact, prudent levels of SES are “punished” by high exposure values and subsequently weaker capital relief, while high levels of SES do offer relative lower exposure values for excess spread and subsequently achieve better capital relief.
In our view, this would represent an incentive for regulatory arbitrage than a restriction. This is due to the fact that the exposure value for future excess spread is limited by the total expected losses according to the asset model approach instead of being based on the amount exceeding expected losses. Increasing excess spread levels would provide substantial credit support to the tranches and therefore strongly improve the risk profile of the tranches. To illustrate this, we would like to refer to the representative transaction on page 25 of the draft RTS. The following table (see attached document) shows that the exposure value does only increase marginally with higher excess spread levels:
As mentioned, this would promote regulatory arbitrage rather than restricting it.
In addition, this would hamper the level playing field between traditional and synthetic transactions, as only synthetic transactions would be affected, while an equivalent traditional transaction would not.
Having in mind the stated aim of both the CRR Quick Fix and the draft RTS, we believe that an approach along the following lines would be more appropriate, economically meaningful and also consistent with the Level-1 text.
By way of derogation from the treatment of traditional securitizations under the CRR one can argue that future excess spread of a securitization covering the expected loss is already risk-weighted under the existing rules of Chapter 5. As indicated in 8 (ii) of the draft RTS in a traditional securitization the originator is giving up future income of securitized assets. This does include the excess spread portion of the future income of assets, which is available to cover losses of the securitized portfolio within the traditional securitization structure in the first place. This credit support in form of future excess spread is of course affecting the risk profile of the tranches and is also considered in pricing.
Subsequently applying the rules of Chapter 5 for the calculation of the risk-weights of tranches implies that credit support provided by future excess spread is already considered in the risk weighting of tranches for traditional securitizations. This is supported by the fact that SEC-SA and SEC-IRBA effectively assign a multiple of the risk-weighted assets of the underlying portfolio to the various tranches according to an exponential function, therefore already considering the expected loss of the underlying portfolio.
So, while in a traditional securitization future income (including the excess spread portion of it) is given up by the originator, in a synthetic securitization the future income (including the excess spread portion of it) stays with the originator, as it is correctly stated in para. 8 (ii) page 6, if no synthetic excess spread is provided. A synthetic securitization therefore is equivalent to a traditional securitization if and only if the originator assigns the excess spread portion of the future income to the structure to cover expected losses in the portfolio in form of a synthetic excess spread. This is both true for the originator’s position in terms of future income as well as in terms of the risk profile of the tranches. In this context we would like to highlight the last paragraph of Art. 248 (e). In our view by way of derogation from the treatment of traditional securitizations it is economically intuitive and well arguable that future synthetic excess spread up to (!) the expected loss is already risk-weighted under the rules of Chapter 5, while any synthetic excess spread portion exceeding the expected loss is not.
We therefore would consider it to be economically meaningful and in line both with the Level-1 text and the intentions of the legislators to focus on future excess spread exceeding the expected loss of the securitized portfolio (plus past/current excess spread which is already considered in the income statement/balance sheet) for calculating the exposure value instead of the future excess spread up to the expected loss. In our view such an interpretation would also better serve the goal to prevent regulatory arbitrage by requiring risk-weighing of excessive future synthetic excess spread.
Finally, the current draft has an excessive undue punitive effect on smaller banks by requiring substantial risk-weighting for prudent levels of synthetic excess spread: due to their smaller loan books both in size and number of positions, securitized portfolios are often longer in average life and therefore more dependent on excess, while a prudent level of excess spread is economically justified. Also securitized portfolios of smaller banks are often more focused towards SMEs, which could reveal a riskier profile, generating higher income, having higher expected losses and again as a result being more dependent on a prudent level of synthetic excess spread.
In case the approach currently taken in the draft RTS is nevertheless maintained, we would also like to note the following.
There is an evident lack of a grandfathering or phase-in provision for existing transactions. Existing transactions structured with excess spread will become uneconomic and even lose their complete RWA relief in the worst case (especially EIF structures include excess spread and a retained first loss tranche that have the risk of falling below 1250% RW with future excess spread inclusion in SEC IRBA formula and thus not be capital deducted anymore but with high RWAs to be considered).
This could trigger a high number of regulatory calls. The consequence will be a big negative effect from loss of capital relief for many European banks in a difficult environment at effectiveness of the proposed paper if the future excess spread calculation is not reconsidered.
This would again be especially burdensome for smaller banks as due to typically smaller transaction sizes and often longer WALs of securitized portfolios sunk costs due to regulatory calls would be higher from a relative perspective.