Response to discussion on the Significant Risk Transfer in Securitisation
Go back
• The percentage of traditional SRT securitisations which were publicly listed transactions compared to private placements.
• The percentage of SRT synthetic transactions which were publicly or privately issued CLNs.
• The percentage of synthetic transactions which were structured as private bilateral guarantees or CDS.
Client-driven activities were not included in the sample. Pre-notifications should not apply to these transactions because they do not have structured features.
In addition, there is too much reputation risk with investors if banks have to call the deal early because it didn’t receive approval in time. In this case, investors may also require some minimum amount of fees and indemnities which globally would lead to increased costs for banks, while no SRT being achieved.
The feedback provided by the regulator before the closing of the transaction should at least highlight:
a. any issues the regulator considers problematic
b. the issues on which they are likely to focus
c. the type of analysis and information that will be required to address any issues cited in b. above.
The FBF also fully agrees with the EBA recommendation that, even for deals that meet the mechanistic tests and therefore do not require specific regulatory approval, originators should be able to expect a no objection from the regulator, as they always have the discretion to reject a deal. It is worth noting that currently in most cases, regulators do not provide feedback in this instance.
EBA proposes that the originator must notify the authority 1 month before the expected start date of the securitization. Today the originator must notify the competent authority 3 months before the expected closing date.
The FBF supports the following solution:
- The originator shall notify the competent authority no later than 45 days before the expected closing date. However this period has to be understood as a discussion period – and should not start once all questions have been answered;
- With respect to §74-b, a preliminary feedback from the supervisory authority shall be provided ahead of the closing of the transaction;
- The competent authority should provide a notification on SRT compliance and non-objection/objection before the expected closing date (i.e. no later than 45 days after the first notification by the originator);
- A non-response can be considered like a no-objection by the regulator;
- When SRT has been agreed or not objected by a competent authority at Group level, there should be a simplified and accelerated process for SRT notification and assessment at local entity level (especially when the local entity is supervised by the same competent authority).
For securitisations structured for capital relief purposes, it is important for the banks to receive an answer from the authority early enough for them to adjust the structure of the securitization to the extent necessary so as to achieve significant risk transfer.
It may also happen that institutions structure securitization transactions (often traditional) in a very short period of time and place them initially for funding purpose (i.e., only place the most senior and low risk tranches at that point) and thus will not be able to comply with the notification period if the junior tranches are placed at a later stage (i.e., after closing) through a separate (and specific) placement. To address this possible dual timing, the SRT process should also contemplate the possibility for institutions to notify SRT after closing, bearing in mind that institutions would not be allowed to consider SRT in their capital requirement calculations before the competent authorities have answered
The deal by deal review by competent authorities provides a sufficient safety net in terms of banking supervision. Should a new SRT notification template be developed, the ECB questionnaire (of 24 March 2016) would provide a good template. Nevertheless, a separate questionnaire should be developed for traditional and synthetic securitisation to avoid confusion. For example, banks sometimes receive questions from competent authorities that indicate that the analyst had misunderstood the transaction under review. That approach would also be consistent with the tight timeframe imposed for reviewing and approving transactions.
If a template is to be developed, it should be different for traditional and synthetic securitizations as said in our answer to question 5. We consider this distinction as useful, as a traditional securitization is usually considered as a standard transaction in contrast to a synthetic securitization. Consequently, all features that are common in synthetic securitizations but not in traditional securitizations or vice-versa are viewed as discrepancies and require further explanation.
Depending on the specifics of the securitization, every trigger may not be relevant. It is unclear if the list given (§92) is meant to be included in all deals as a minimum requirement:
- Some triggers are not applicable to all asset types. For example, the granularity test would not be relevant to granular portfolios where no debtor or group of debtors would ever be expected to reach a meaningful percentage of the pool.
- Some triggers are not applicable to all banks. For example, the expected loss trigger is not computable for a bank under the standard approach.
As banks often enter into these transactions for regulatory capital relief, an interesting idea would be to link the trigger to the parameters that are used to assess the SRT. For example, the amortisation switch could be triggered by an increase of the Kirb or in the event that a mezzanine tranche becomes an equity tranche.
Finally, the FBF recommends providing some flexibility to adapt and/or omit some of these triggers when not relevant or not appropriate.
There is no clear argument about the reason why the first call date should be so far out in the life of the transaction.
A rule forbidding exercising the call option before “the end of the replenishment period plus the initial weighted average life” would be too restrictive, as the maturity mismatch constraint strongly encourages the originator to replenish with shorter loans maturing before the synthetic securitisation maturity date. For this reason, for most securitisations, at the end of the replenishment period, the residual weighted average life (“WAL”) will be significantly shorter than the initial WAL. Because of the restrictive maturity mismatch rule, it is not common that securitisations are replenished with loans that mature after the initial loans. Hence it is uncommon that the WAL of the portfolio increases during the replenishment period.
In our view, the rule should be that the call option cannot be exercised before “the end of the replenishment period plus the weighted average life of the portfolio at the end of the replenishment period”.
However, as the WAL of the portfolio at the end of the replenishment period is not known at the closing date, it might be difficult to include such a rule in the contractual documentation.
The FBF supports the following solutions:
Solution 1: The originator must be able to demonstrate that it exercises the call in its own interest and is not doing a favour to the investor.
Or
Solution 2: The time call can be exercised subject to prior approval of the competent authority.
Referring to the Basel framework approach, time calls are acceptable as long as there is no structural incentive, for the originator, to exercise them. As such, certain features pointing to a structural incentive (e.g. step up premiums) should be disclosed by the originator in its submission, reviewed by the competent authority and would probably be refused.
But, if the credit risk diminishes substantially (for example due to a reduction in the average PD, LDG, and/or IRBA maturity of the pool, this is typically reflected when a KIRB is lower than the one at inception) the cost of the hedging becomes too expensive relative to the value of the hedging. This situation could happen before the proposed non-call period. It doesn’t make sense to impose this non-call period if the exercise of the call benefits to the originator. In that case, exercising such call does not arise from an incentive to call or a support to the investor.
Other comments:
• 102 a (II) : it should be possible to include call options regarding economic efficiency such as rating decisions from the credit rating agencies to the extent that they may have an impact on the cost of capital relief and SRT.
• There should be no differences for time calls between traditional and synthetic securitizations as there are no differences in terms of capital relief.
First we would like to underline that the accounting deconsolidation does not rely only on the transfer of risk but rather on the control being surrendered over, and the risks and rewards being transferred on the assets. Moreover, EBA recognises in the discussion paper (§429) that the “accounting derecognition is thus very difficult to achieve, and only a very limited number of transaction have managed to derecognise the securitised assets from their balance sheet”. This EBA statement is thus somehow contradictory with the comments made under §119.
Secondly, we think that the excess spread that can be extracted from the transaction to the benefit of the originator relates only to the revenues of the assets and not to their risk: the institution has transferred their credit risk and will take profit of the revenues of these assets but this right on future excess spread does not expose the institution to any credit risk (unless this right is valuated .
Nevertheless we recognise that if the right to future excess spread is valuated in the balance sheet of the originator, this value should be subject to a risk weight of 1250% for the SRT measurement and Pilar 1 purposes.
Contractually fixed excess spread commitment
As rightly pointed out in the EBA consultation paper in §106, excess spread for Synthetic Securitisations can be:
- A “Variable Excess Spread”[§106.b] which would be computed in a way similar to a Traditional Securitisation and which would consequently be dependent on the performance of the underlying portfolio and also on the effective yield of the securitised portfolio
- A “Fixed Excess Spread”[§106.a], which would be set a certain percentage of the portfolio balance, irrespective of the effective yield of the securitised portfolio and of the potential costs of the structure and of the most senior tranches of the securitised portfolio.
The EBA proposal in [§123.a] restricts the excess spread for synthetic securitisation to a Fixed Excess Spread, and as there is possibly no ground on the value of this Fixed Excess Spread, the EBA considers this should be treated as an additional protection granted to the investors by the Originating Bank.
The FBF members strongly advocate for the possibility to allow a Variable Excess Spread to the extent it is determined to replicate the equivalent of an excess spread in a Traditional Securitisation:
- If it was to be determined in accordance with what would be done for a Traditional Securitisation, there would be no concern in [§120.a] in relation to “the quantum of synthetic excess spread”(which seems more to be a point for a Fixed Excess Spread).
- Regarding the EBA concern expressed in [§120.b] on the “Calculation of Variable Synthetic Excess Spread”, the FBF members note that:
o The determination of servicing costs would not be different from that of a Traditional Securitisation for which the Originator generally remains the servicer and has to determine a fee payable by the SPPE. In addition, the servicing costs of Traditional Securitisation backed by similar assets can serve as a reference point to determine such servicing costs.
o In relation to the fact that the originator generally keeps the tranches senior to the protected tranche(s) on its balance sheet : the equivalent cost of funding of these retained tranches can be determined by references to the senior/high mezzanine tranches for Traditional Securitisation backed by the same asset class or similar assets (the FBF members note in that respect that a significant part of the market for Traditional Securitisation related to the senior tranches of securitisations).
Also Variable Excess Spread – to the extent it is computed in an equivalent way as for Traditional Securitisiations (ie dependent on the effective performance of the portfolio, on its effective yield and on servicing/senior costs) – shall not attract any risk-weight in order to be consistent with the treatment of such excess spread if there was no securitisation in place (if there is no securitisation in place, there is no risk weight for future and uncertain excess margin on a portfolio).
Finally, excess spread is vital for securitisations of granular asset classes (eg. Consumer, retail or SME exposures) where the granularity of the portfolio means that the portfolio will effectively suffer from principal losses due to the high number of securitised exposures. As such, if only Fixed Excess Spread (as defined in the EBA proposal and with a capital requirement attached to it as proposed by the EBA) was to be allowed, this would mean that this would render the use of Synthetic Securitisation ineffective for such portfolios: SRT would only be possible with Traditional Securitisations format where Variable Excess Spread could be used.
Trapping of excess spread
It is unclear why the ‘trapping mechanism’ is preferred to the ‘use-it-or-lose-it mechanism’, while the ‘trapping mechanism’ achieves less risk transfer than the latter.
We don’t understand why the ‘trapping mechanism’ is preferred to the ‘use-it-or-lose-it mechanism’, while the ‘trapping mechanism’ achieves less risk transfer than the latter. In particular the trap mechanism will offer a better protection to the investors as the unused protection provided by the excess spread on a given period will be reported for the following periods during the trapping period, while the ‘use it or lose it’ will always cap the amount of credit protection granted by the originator for a given period.
Moreover, in the trap mechanism the cumulated excess spread will progressively create a credit enhancement provided by the originator that is not compatible with the objective of capital relief since this cumulated excess spread will constitute a first loss subject to 1250% RW.
Thus we think that the EBA should allow both features (use-it-or-lose-it and trap).
Generally speaking, if the committed Fixed Excess Spread for given period cannot be justified by the actual yield on the portfolio/liabilities costs or is fixed whatever this yield/these liabilities costs, then it should be accounted as additional protection for this period as this is protection in addition to what the portfolio can generate on this period. This additional protection will translate into higher attachment point for the protected tranches – hence the excess spread commitment for a given period will be considered as a first loss retained tranche and likely be subject to 1250% RW/Capital Deduction.
The same would hold for excess spread which would have already been trapped (as this would be equivalent to a cash reserve).
However, the proposal [§123.a III] – which relates to trapped Fixed Excess Spread for a 1-year period - mentions a 1250%RW/Capital Deduction not only for the trapped excess spread but for the future component of the Fixed Excess Spread.
* If the Fixed Excess Spread for the future periods is a fixed percentage of the portfolio outstanding balance of the portfolio at the beginning of each observation period (e.g. 30 bppa applied to the portfolio outstanding balance at the beginning of each quarterly period), it may be difficult to have the amount for the 1-year period accounted as a first loss tranche (presumably subject to 1250%/capital deduction) as the effective amount of Fixed Excess Spread may not be known as the portfolio will amortise.
* If the Fixed Excess Spread for the future periods is simply a fixed amount set in advance of the 1-year period whatever the performance or amortisation of the portfolio (e.g.30bbpa applied to the portfolio outstanding balance at the beginning of each calendar year), this is then equivalent to a non-amortising First Loss retained tranche (which would be presumably subject to 1250%/capital deduction). However, in this case, having such a set-up for this Fixed Excess Spread (equivalent to a fixed first loss) may prevent the capital relief for deals in amortisation phase for example.
For FBF members’ position on the type of excess spread which should be allowed for Synthetic Securitisations, please refer to answer to question 11. In this answer, the FBF members provide also their position of the treatment of Variable Excess Spread for Synthetic Securitisations – which shall be treated as excess spread for Traditional Securitisations and attract no RW as it is the case for any unsecuritised portfolio (no RW in relation to future margin).
As mentioned in the answer to question 12, there may be different ways of computing the excess spread commitment:
• We assume the case in this answer where it is computed as a fixed amount of the portfolio outstanding balance at the beginning of each observation period (for the other case where the Fixed Excess Spread is a fixed amount set at the beginning of each yearly period whatever the performance and amortisation of the portfolio, please refer to answer to question 12 where the Fixed Excess Spread can be considered as a unprotected first loss tranche).
• In this case, If the portfolio outstanding balance excludes the exposures which have defaulted, this means that the committed excess spread depends on the future performance of the securitised portfolio and hence exists in the future to the extent that the portfolio is performing and has not amortised.
As there is no certainty on this committed Fixed Excess Spread, the FBF members are on the opinion that the unrealised/uncommitted component of Fixed Excess Spread should not attract any RW.
a. Allowing the investment of cash proceeds in eligible investments held at an independent custodian. According to the latest EBA guidance, this structure would be prohibited for STS SME synthetics and presumably other asset classes if STS benefits are extended to other synthetics.
b. Allowing the cash to be deposited with an originator segregated from the bankruptcy estate of the originator through deposits in a bankruptcy remote custodian account, investments in eligible investments, repos and other similar structures.
c. Downgrade triggers requiring the transfer of deposited funds to another bank or its investment in eligible collateral. The EBA should clarify that such clauses are not problematic for SRT, and that movements of funds due to such clauses would not cause the loss of STS status of a deal.
In some synthetic securitizations of certain asset classes, removing termination in case of originator bankruptcy could have a material impact on rated transactions, as ratings are generally dependent on servicing continuity. This would also be a concern for investors. For example, the performance of larger concentrated assets such as commercial real estate or certain commercial loans, such as SMEs may be highly correlated to the financial health of the originator to the extent servicing staff were made redundant. Accordingly any ban on termination clauses involving originator bankruptcy must take these concerns into account. Investors would have to be given the right to terminate in a bankruptcy situation to the extent that servicing arrangements (and indeed investor reporting obligations) were not carried out satisfactorily.
- The EBA proposes to use the latest EU wide stress tests parameters as stress scenario on the PD & LGD to the underlying exposures
- We understand that institutions would compare the amount of losses transfered in this scenario of stress to the amount of RWA transfered.
- The amount of RWA represent an amount of unexpected loss which is defined, under the IRB approach, by the RW formula provided by the Basel committee.
- This Unexpected Loss represent a stress scenario on the PD that is much higher than the stress scenario used for the EU stress test :
o For instance, the implicit stress for a corporate with a current PD of 0,10% under the IRB framework is a stressed PD of 3,42% while the EU stress test results in a stressed PD of 0,16%.
o A PD of 1% result in a stressed PD of respectively 14% (IRB) and 1,61% (EU stress test).
- Thus we think that if it relies on the EU stress test the self-assessment of SRT could by construction result in an absence of risk transfer since banks will compare 2 scenarios (IRB vs EU stress test) whose amplitude have no common measure.
Notwithstanding this, with regards to the whole process, our understanding of the EBA’s proposition is that in order to achieve SRT:
- All securitization transactions will need to pass the new Enhanced Quantitative Tests (EQT) - option 1 or 2.
- Synthetic transactions and complex traditional securitizations will need to pass the new enhanced quantitative tests and the new Quantitative Self-Assessment of Risk Transfer (QSART).
- Simple and traditional securitization transactions will need to pass only the new EQT to achieve SRT.
This approach has two shortcomings:
- The quantification of retained and transferred risk in the new EQT relies on regulatory parameters such as EL and own fund requirements which may in some cases be disconnected from the real level of risk of the exposures, especially for banks using the standardised approach and for NPL portfolios.
- The EQT and the QSART have the same purpose. Both intend to assess whether the reduction in own funds requirements is justified by a commensurate transfer of credit risk. Therefore having to pass both tests for synthetic or complex transactions would be redundant.
We propose the following changes to the approach:
- The issuers of simple and traditional securitisation should be allowed to achieve SRT by using the QSART instead of the new EQT. However, the possibility for simple and traditional transactions to achieve SRT by using the EQT should be maintained.
- As the QSART relies on a more accurate measurement of risk than the EQT, transactions which have used the QSART to prove the commensurateness of the risk transfer should not need to pass the EQT. This should be applicable to any transaction regardless of their complexity and their nature (synthetic or traditional).
- These principles should be applicable to NPL transactions as well.
1. In the first illustration below, 80% of the EL and UL risk is sold to investors, which meets the substantive requirements of the EBA guidance. However, in the first example, the senior retained tranche has been divided into senior and mezzanine tranches (which is the most efficient structure under the SEC-ERBA and SEC-SA approach). Because the mechanistic test states that if mezzanine tranches are present, the originator must sell >50% of the risk weight of the mezzanine tranches, the mechanistic test fails. Nevertheless, in the second example where no mezzanine tranches are present, the first loss test is passed. The first example fails, even though substantively the same amount of risk (which exceeds EL & UL) has been transferred to investors. We believe this is an unintended consequence of the mechanistic tests which should be corrected, if the rest of the EBA guidance is ultimately agreed requiring that a significant portion of the EL and UL risk has been transferred.
[see illustration in the attached file]
2. The minimum first loss test requiring that the first loss tranche cover 100% of EL and 2/3 of UL should be modified to take into account a scenario such as the one illustrated below. In this graph, we see that the originator has transferred 100% of the EL and UL to investors. The first loss tranche does not meet the first loss test, but in substance, the combination of the mezzanine and first loss tranches sold is sufficient to transfer 100% of EL and 2/3 of UL. If not worded correctly, the first loss tranche would inadvertently disallow the illustrated transaction.
[see illustration in the attached file]
3. The revised definition of mezzanine tranches, which excludes mezzanine tranches weighted 1250% exacerbates the problem described above. Under the newly adopted capital calibrations, it is likely that many mezzanine tranches will be weighted 1250% (such as in the case of IRB consumer loans in which the SEC-IRBA approach does not recognise the benefit of excess spread). Accordingly, in the above example, if the most subordinate mezzanine tranche is weighted 1250%, the mezzanine test would fail, even though the originator has transferred risk equal to EL + 2/3 UL.
The test should be restated or clarified to take into account transactions in which a sufficiently large portion of mezzanine risk is sold to cover UL, and where the attachment point of the sold mezzanine tranches encompasses the portion of EL not covered by a thin first loss tranche. In most cases we find that the optimal cost of protection and RWA reduction can be achieved if a small first loss tranche, is retained, while transferring a sufficient amount of mezzanine tranches to cover the rest of EL and most of UL. To prevent the abuse that this test is intended to cover, the EBA could specify that the mezzanine test will not cause a transaction to fail the minimum required for SRT if the originator transfers any combination of first and mezzanine tranches with a thickness equal to EL + 2/3 UL. Another alternative would be to require banks using the mezzanine test to sell a greater percentage of mezzanine tranches than 50%, such that the percentage sold above 50% equals the amount of EL not covered by the first loss tranche.
With respect to the new CRR securitization framework, senior risk weights tend to increase dramatically whereas mezzanine risk weights tend to be rather stable and first loss tranches remain to be deducted/with 1250% risk weight. EL and provisioning of the underlying portfolio remain however unchanged.
From the originator’s perspective, the focus is to achieve economically efficient SRT. With senior risk weights being most impacted from the new regulation, originators might tend to structure senior tranches in a way to keep risk weights on retained senior positions as low as possible. This will likely be achieved by higher amounts of subordinated tranche notional (incl. Excess Spread). In this respect, we rather expect senior tranche sizes to shrink and mezzanine as well as first loss tranche sizes (incl. Excess Spread) to increase. However, a more likely effect in most cases is that banks will cease doing SRT deals because it is likely that the incremental cost of selling additional tranches will make the transaction uneconomic compared to the bank’s cost of capital, and the sale of such additional tranches might cause the bank to fail the cost of credit protection test as proposed by the EBA.
From the investor’s perspective, the focus is to take on an investment with an attractive risk/reward profile. Assuming the investor is not regulated under the CRR, the new regulation is not key in the assessment of risks and rewards. The focus is rather on (economic) expected and unexpected losses of the underlying portfolio in order to assess the risk of the investment. We expect the impact of the new CRR regulation on the investor´s minimum tranche thickness requirements to be marginal.
However, those investors most interested in first and second loss tranches have yield requirements higher than the weighted average cost of protection required by originators for a deal to make economic sense. If the above-mentioned anomaly in the first loss and mezzanine tranches is not corrected, it will be difficult, if not impossible, for originators to sell a thick enough portion of tranches at an economic level.
1. In many cases the new definition of mezzanine, excluding mezzanine tranches which are risk weighted 1250% causes the deal to fail the mezzanine test, even in deals which have transferred tranches equal to EL+2/3UL. The likelihood of mezzanine tranches being risk weighted 1250% is much greater under the new CRR approaches.
2. As per our responses to questions 19, 20 and 21 above, the minimum first loss tranche thickness test inadvertently causes several of our transactions to fail, even though we have transferred a total amount of tranches exceeding EL+2/3UL.
We would like to mention another inconsistency that particularly seems to occur to synthetic transactions which are structured to pass the mezzanine test and where the originator is obliged to apply SEC-IRBA.
In case the first loss tranche plus excess spread are retained by the originator we acknowledge the introduction of the minimum first loss piece thickness test as a helpful measure to ensure that the lifetime EL of a portfolio is covered by the originator whereas its UL is transferred. Given this intention, it is unclear why the lifetime EL is considered in ratio 2 of the commensurate risk transfer test. In fact, sections 216 and 218 reveal that ratio 2 of aims at measuring “the risk transferred to the third parties” obtained “as the share of losses transferred to third parties for the life of the transaction”. We understand that the commensurateness test is not solely directed towards transactions that intend to pass the mezzanine SRT test. However, it is our understanding that the test of commensurate risk transfer contradicts the minimum tranche thickness test particularly in case the mezzanine SRT test applies. This is due to the fact that the latter requires the lifetime EL not to be part of the transferred risk whereas the commensurateness test considers it to be transferred.
Furthermore, the reason why the commensurate risk transfer test compares a 1y measure, ratio 1, with a lifetime measure, ratio 2 is not clear. Even though it is clear why comparing capital savings to transferred risk, it remains questionable why actually basing these compared figures on different time horizons.
Also it seems that Ratio 2 takes into account the lifetime excess spread: if the excess spread is committed and trapped, indeed this should be considered as a retained position – however as detailed in earlier questions, unearned excess spread or un-trapped excess spread should not be considered as a retained position
However, it is important to note that for synthetic securitizations where synthetic excess spread of the amount of the EL is trapped on an annual basis, the 1st Excess Spread payment that is trapped towards the end of the 1st year should not be considered in the calculation of Condition 1 since it serves as a buffer of the EL of the underlying portfolio that materializes during the 2nd year of the transaction.
In analysing the impact of this test on banks existing SRT positions, we note that, in many cases, the increased capital on senior retained tranches required by the new capital calibrations under the revised CRR causes our SRT transactions to fail, even though the positions sold to investors exceed EL+2/3UL, which is the minimum suggested by the EBA. The effect of the new calibrations is to require an originator to sell a much larger portion of first loss and/or mezzanine tranches in order to reduce the risk weight of the retained senior tranches below the threshold required by the test. For any particular transaction, some or all of the following negative impacts then may apply:
a. It increases the cost of protection above the bank’s cost of capital, making the deal uneconomic to execute.
b. It causes the total cost of protection to exceed the EBA guidelines on the cost of credit protection.
c. It requires the originator to sell total tranche thicknesses well in excess of EL+2/3UL, which makes such test redundant or inappropriate.
• Generally, the key issue regarding NPL transfer is that the NPL are not anymore included on the balance sheet of the originator and hence it obtains a decrease of its NPL rate.
• The SRT and regulatory capital relief are usually not key drivers in our sense for such NPL transfers.
• Consequently, the question of who will be in charge of the servicing, the originator or an independent third party manager, is critical for the structuring of NPL securitizations. As a matter of fact, when the originator of the assets remains the manager/servicer after the assets have been “transferred” to the SSPE, the assets can usually not be derecognized or the SSPE be deconsolidated in application of the IFRS accounting rules (control test).
• Another point of attention is the diverging views that the originator and the investors can have on the expected cash flows.
• Barriers to entry for NPL investors: in accordance with French regulation, loan servicers have to hold a licence to acquire and service French loans. It remains a hurdle for the development of the NPL market. Moreover, to acquire large NPL portfolios, buyers will have to acquire and potentially develop existing dedicated platforms that have sufficient trained and knowledgeable staff and the technical capacity to service large numbers of loans.
• Limitations on the transferability of loans: banks wishing to sell any loan portfolios are required to remove the loans which include a clause of non-transferability into the underlying contract. Such a clause continues to be in force once a loan becomes non-performing.
• Disclosure limitations during certain legal proceedings: banks are not allowed to disclose to potential buyers that a mandate or conciliation procedure is underway, which complicates the sale of any NPLs affected by such a legal proceeding.
In addition, public sector/state guarantee on NPL securitisations available in other countries like Italy with efficient payment mechanisms could facilitate the distribution / placement of the NPL securitisation transactions. This guarantee could cover the senior notes and/or as the case may be, the mezzanine notes.
Nevertheless, we think that the future EBA guidelines will need to take into account future regulation from the EC and the guidance from the ECB on the prudential provisioning of NPL. Indeed the application of the proposal made by the EBA under figure 25 of the discussion paper to a NPL portfolio would result in significant capital relief if the future securitization framework does not take in account the prudential provisioning and thus could lead to significant incentive for securitization and regulatory arbitrage.
We also note that there is a wide discrepancy between the results of the SEC-IRBA and the SEC-SA on the Risk Weights of the different classes of notes of the NPL securitisation example (Fig. 25).
In addition, the RWA of the junior notes seems at a first glance relatively low compare to the RWA of other mezzanine or junior notes of usual securitisations. On the contrary, the RW on all the notes with the SEC SA are extremely high and in particular on the senior tranche.
The overly conservatism of the SEC-SA for NPL pools stems in our view from the way the ‘W’ factor is taken into account in the SEC-SA. This issue is not specific to NPL but it is amplified in the case of NPL pools. It was already identified with a simple correction proposed in a paper called: “How to revive the European Securitisation Market: a Proposal for a European SSFA” [*] by Georges Duponcheele Alexandre Linden and William Perraudin published in November 2014. Indeed the paper mentions: “It is to be noted that this official version of the SA in the BCBS 269 proposals contains a conceptual error in that it includes the delinquent assets (via Ka) in Kp. This in effect means that the capital associated with provisions for the delinquent assets is allocated to the senior tranches via the exponential function and the parameter a.
To correct this conceptual error, Kp should simply equal pool capital for performing loans and should not be a function of W. The delinquent assets should only affect the 100% capital charge threshold KT. It is important to distinguish between KT and KP to avoid making errors.”
We therefore reiterate the need to correct the way the W factor is taken into account in the SEC-SA as detailed in the extract above.
[*] https://www.researchgate.net/publication/279960182_How_to_Revive_the_European_Securitisation_Market_a_Proposal_for_a_European_SSFA_Alexandre_Linden_William_Perraudin_BNP_Paribas_BNP_Paribas_Risk_Control_Limited
Moreover, SRT is not the main driver for NPL securitizations; major drivers are the search for senior refinancing or decrease of the balance sheet NPL rate (see our answer to question 27)
Question 1: Does the data on synthetic and traditional SRT securitisation transactions correspond with your assessment of SRT market activity in the EU? Do you have any observations on these data?
It seems that some important aspects of securitisation transactions were not analysed:• The percentage of traditional SRT securitisations which were publicly listed transactions compared to private placements.
• The percentage of SRT synthetic transactions which were publicly or privately issued CLNs.
• The percentage of synthetic transactions which were structured as private bilateral guarantees or CDS.
Client-driven activities were not included in the sample. Pre-notifications should not apply to these transactions because they do not have structured features.
Question 2: Are you aware of any material supervisory practices that have not been covered in the EBA analysis?
First of all, the FBF fully shares the opinion (paragraph 70) that lengthy feedback procedures may result in a high degree of uncertainty for the market and that it is highly important to receive feedback from the regulatory authority sufficiently ahead of the closing of the transaction which is currently not ensured. SRT decisions may have important consequences for the structuring of the transaction and banks should have some time to adapt the structural features of the transaction as a consequence of the regulator’s feedback. It is also noteworthy that the transaction costs may be too high to incur if ultimately the SRT deal is rejected.In addition, there is too much reputation risk with investors if banks have to call the deal early because it didn’t receive approval in time. In this case, investors may also require some minimum amount of fees and indemnities which globally would lead to increased costs for banks, while no SRT being achieved.
The feedback provided by the regulator before the closing of the transaction should at least highlight:
a. any issues the regulator considers problematic
b. the issues on which they are likely to focus
c. the type of analysis and information that will be required to address any issues cited in b. above.
The FBF also fully agrees with the EBA recommendation that, even for deals that meet the mechanistic tests and therefore do not require specific regulatory approval, originators should be able to expect a no objection from the regulator, as they always have the discretion to reject a deal. It is worth noting that currently in most cases, regulators do not provide feedback in this instance.
Question 3: What are your views on the proposals on the assessment process set out above? Are any other changes necessary to further improve the process?
A this stage, it is not clear how this requirement would apply to portfolio supervised under a standard approach.EBA proposes that the originator must notify the authority 1 month before the expected start date of the securitization. Today the originator must notify the competent authority 3 months before the expected closing date.
The FBF supports the following solution:
- The originator shall notify the competent authority no later than 45 days before the expected closing date. However this period has to be understood as a discussion period – and should not start once all questions have been answered;
- With respect to §74-b, a preliminary feedback from the supervisory authority shall be provided ahead of the closing of the transaction;
- The competent authority should provide a notification on SRT compliance and non-objection/objection before the expected closing date (i.e. no later than 45 days after the first notification by the originator);
- A non-response can be considered like a no-objection by the regulator;
- When SRT has been agreed or not objected by a competent authority at Group level, there should be a simplified and accelerated process for SRT notification and assessment at local entity level (especially when the local entity is supervised by the same competent authority).
For securitisations structured for capital relief purposes, it is important for the banks to receive an answer from the authority early enough for them to adjust the structure of the securitization to the extent necessary so as to achieve significant risk transfer.
It may also happen that institutions structure securitization transactions (often traditional) in a very short period of time and place them initially for funding purpose (i.e., only place the most senior and low risk tranches at that point) and thus will not be able to comply with the notification period if the junior tranches are placed at a later stage (i.e., after closing) through a separate (and specific) placement. To address this possible dual timing, the SRT process should also contemplate the possibility for institutions to notify SRT after closing, bearing in mind that institutions would not be allowed to consider SRT in their capital requirement calculations before the competent authorities have answered
Question 4: Could you provide suggestions as to whether and how the template for SRT notification by the competent authority to EBA provided in Annex I of the EBA Guidelines should be amended to reflect the new EU securitisation framework and the STS securitisation product?
FBF answer: No change should be made to the template – except adding a field corresponding to the STS nature of the transaction (if this is the case).Question 5: Should a standardised SRT notification template be developed, for submission by originators to competent authorities, in order to facilitate the SRT assessment process? If yes, should this template be different for traditional and synthetic securitisation? (Please provide examples of templates, as appropriate).
It is important to strike the right balance between standardization and recognition of the diversity of situations: asset classes, geographies, local laws makes necessary imply retaining some level of flexibility.The deal by deal review by competent authorities provides a sufficient safety net in terms of banking supervision. Should a new SRT notification template be developed, the ECB questionnaire (of 24 March 2016) would provide a good template. Nevertheless, a separate questionnaire should be developed for traditional and synthetic securitisation to avoid confusion. For example, banks sometimes receive questions from competent authorities that indicate that the analyst had misunderstood the transaction under review. That approach would also be consistent with the tight timeframe imposed for reviewing and approving transactions.
Question 6: Could you provide suggestions as to how a template for monitoring SRT compliance should look like (e.g. by potential amendments of the current COREP framework)?
The multiplication of ad hoc reporting by institutions creates operational risk and unnecessary costs. We suggest the EBA to incorporate the SRT monitoring reporting into the COREP framework.If a template is to be developed, it should be different for traditional and synthetic securitizations as said in our answer to question 5. We consider this distinction as useful, as a traditional securitization is usually considered as a standard transaction in contrast to a synthetic securitization. Consequently, all features that are common in synthetic securitizations but not in traditional securitizations or vice-versa are viewed as discrepancies and require further explanation.
Question 7: Do you agree with the assessment of the SRT implications of all the identified structural features? Are any material aspects missing from this representation?
We ask for retaining flexibility to use alternative amortisation structures such as a hybrid approach that features sequential amortisation until the protected tranche reaches a certain percentage threshold, after which point the amortisation switches to pro-rata amortization.Question 8: Do you agree with the proposed safeguards related to the use of pro-rata amortisation?
The contractual triggers that switch from pro-rata amortisation to a sequential amortisation would significantly increase the costs of the securitisations. Moreover, having multiple constraints leading to a definitive switch to pro rata may in practice result in not allowing pro rata amortization for some asset classes.Depending on the specifics of the securitization, every trigger may not be relevant. It is unclear if the list given (§92) is meant to be included in all deals as a minimum requirement:
- Some triggers are not applicable to all asset types. For example, the granularity test would not be relevant to granular portfolios where no debtor or group of debtors would ever be expected to reach a meaningful percentage of the pool.
- Some triggers are not applicable to all banks. For example, the expected loss trigger is not computable for a bank under the standard approach.
As banks often enter into these transactions for regulatory capital relief, an interesting idea would be to link the trigger to the parameters that are used to assess the SRT. For example, the amortisation switch could be triggered by an increase of the Kirb or in the event that a mezzanine tranche becomes an equity tranche.
Finally, the FBF recommends providing some flexibility to adapt and/or omit some of these triggers when not relevant or not appropriate.
Question 9: Do you agree with the proposed safeguards related to the use of time calls? Do you agree with the different approach to time calls in traditional vs. synthetic transactions?
We fail to understand the theoretical framework that justifies the proposed constraints on time calls.There is no clear argument about the reason why the first call date should be so far out in the life of the transaction.
A rule forbidding exercising the call option before “the end of the replenishment period plus the initial weighted average life” would be too restrictive, as the maturity mismatch constraint strongly encourages the originator to replenish with shorter loans maturing before the synthetic securitisation maturity date. For this reason, for most securitisations, at the end of the replenishment period, the residual weighted average life (“WAL”) will be significantly shorter than the initial WAL. Because of the restrictive maturity mismatch rule, it is not common that securitisations are replenished with loans that mature after the initial loans. Hence it is uncommon that the WAL of the portfolio increases during the replenishment period.
In our view, the rule should be that the call option cannot be exercised before “the end of the replenishment period plus the weighted average life of the portfolio at the end of the replenishment period”.
However, as the WAL of the portfolio at the end of the replenishment period is not known at the closing date, it might be difficult to include such a rule in the contractual documentation.
The FBF supports the following solutions:
Solution 1: The originator must be able to demonstrate that it exercises the call in its own interest and is not doing a favour to the investor.
Or
Solution 2: The time call can be exercised subject to prior approval of the competent authority.
Referring to the Basel framework approach, time calls are acceptable as long as there is no structural incentive, for the originator, to exercise them. As such, certain features pointing to a structural incentive (e.g. step up premiums) should be disclosed by the originator in its submission, reviewed by the competent authority and would probably be refused.
But, if the credit risk diminishes substantially (for example due to a reduction in the average PD, LDG, and/or IRBA maturity of the pool, this is typically reflected when a KIRB is lower than the one at inception) the cost of the hedging becomes too expensive relative to the value of the hedging. This situation could happen before the proposed non-call period. It doesn’t make sense to impose this non-call period if the exercise of the call benefits to the originator. In that case, exercising such call does not arise from an incentive to call or a support to the investor.
Other comments:
• 102 a (II) : it should be possible to include call options regarding economic efficiency such as rating decisions from the credit rating agencies to the extent that they may have an impact on the cost of capital relief and SRT.
• There should be no differences for time calls between traditional and synthetic securitizations as there are no differences in terms of capital relief.
Question 10: Do you agree with the proposed safeguards on the use of excess spread in traditional securitisation?
The EBA in its discussion paper indicates that when the excess spread can be extracted from the transaction to the benefit of the originator and that there is no accounting deconsolidation, then the SRT should be assessed with particular focus.First we would like to underline that the accounting deconsolidation does not rely only on the transfer of risk but rather on the control being surrendered over, and the risks and rewards being transferred on the assets. Moreover, EBA recognises in the discussion paper (§429) that the “accounting derecognition is thus very difficult to achieve, and only a very limited number of transaction have managed to derecognise the securitised assets from their balance sheet”. This EBA statement is thus somehow contradictory with the comments made under §119.
Secondly, we think that the excess spread that can be extracted from the transaction to the benefit of the originator relates only to the revenues of the assets and not to their risk: the institution has transferred their credit risk and will take profit of the revenues of these assets but this right on future excess spread does not expose the institution to any credit risk (unless this right is valuated .
Nevertheless we recognise that if the right to future excess spread is valuated in the balance sheet of the originator, this value should be subject to a risk weight of 1250% for the SRT measurement and Pilar 1 purposes.
Question 11: Do you agree with the proposed safeguards constraining the use of excess spread in synthetic securitisation? In particular, do you agree with: a. The proposal of only allowing a contractually fixed (pre-determined) excess spread commitment in synthetic transactions? b. The proposal to only allow a ‘trap’ excess spread allocation mechanism in synthetic transactions?
The FBF members strongly disagree with the proposed limitation of “excess spread” for synthetic securitisations, especially with the proposal to only allow (i) a contractually fixed excess spread commitment in a synthetic securitisation and (b) a trapping mechanism of excess spread (instead of a use-it-or-lose it).Contractually fixed excess spread commitment
As rightly pointed out in the EBA consultation paper in §106, excess spread for Synthetic Securitisations can be:
- A “Variable Excess Spread”[§106.b] which would be computed in a way similar to a Traditional Securitisation and which would consequently be dependent on the performance of the underlying portfolio and also on the effective yield of the securitised portfolio
- A “Fixed Excess Spread”[§106.a], which would be set a certain percentage of the portfolio balance, irrespective of the effective yield of the securitised portfolio and of the potential costs of the structure and of the most senior tranches of the securitised portfolio.
The EBA proposal in [§123.a] restricts the excess spread for synthetic securitisation to a Fixed Excess Spread, and as there is possibly no ground on the value of this Fixed Excess Spread, the EBA considers this should be treated as an additional protection granted to the investors by the Originating Bank.
The FBF members strongly advocate for the possibility to allow a Variable Excess Spread to the extent it is determined to replicate the equivalent of an excess spread in a Traditional Securitisation:
- If it was to be determined in accordance with what would be done for a Traditional Securitisation, there would be no concern in [§120.a] in relation to “the quantum of synthetic excess spread”(which seems more to be a point for a Fixed Excess Spread).
- Regarding the EBA concern expressed in [§120.b] on the “Calculation of Variable Synthetic Excess Spread”, the FBF members note that:
o The determination of servicing costs would not be different from that of a Traditional Securitisation for which the Originator generally remains the servicer and has to determine a fee payable by the SPPE. In addition, the servicing costs of Traditional Securitisation backed by similar assets can serve as a reference point to determine such servicing costs.
o In relation to the fact that the originator generally keeps the tranches senior to the protected tranche(s) on its balance sheet : the equivalent cost of funding of these retained tranches can be determined by references to the senior/high mezzanine tranches for Traditional Securitisation backed by the same asset class or similar assets (the FBF members note in that respect that a significant part of the market for Traditional Securitisation related to the senior tranches of securitisations).
Also Variable Excess Spread – to the extent it is computed in an equivalent way as for Traditional Securitisiations (ie dependent on the effective performance of the portfolio, on its effective yield and on servicing/senior costs) – shall not attract any risk-weight in order to be consistent with the treatment of such excess spread if there was no securitisation in place (if there is no securitisation in place, there is no risk weight for future and uncertain excess margin on a portfolio).
Finally, excess spread is vital for securitisations of granular asset classes (eg. Consumer, retail or SME exposures) where the granularity of the portfolio means that the portfolio will effectively suffer from principal losses due to the high number of securitised exposures. As such, if only Fixed Excess Spread (as defined in the EBA proposal and with a capital requirement attached to it as proposed by the EBA) was to be allowed, this would mean that this would render the use of Synthetic Securitisation ineffective for such portfolios: SRT would only be possible with Traditional Securitisations format where Variable Excess Spread could be used.
Trapping of excess spread
It is unclear why the ‘trapping mechanism’ is preferred to the ‘use-it-or-lose-it mechanism’, while the ‘trapping mechanism’ achieves less risk transfer than the latter.
We don’t understand why the ‘trapping mechanism’ is preferred to the ‘use-it-or-lose-it mechanism’, while the ‘trapping mechanism’ achieves less risk transfer than the latter. In particular the trap mechanism will offer a better protection to the investors as the unused protection provided by the excess spread on a given period will be reported for the following periods during the trapping period, while the ‘use it or lose it’ will always cap the amount of credit protection granted by the originator for a given period.
Moreover, in the trap mechanism the cumulated excess spread will progressively create a credit enhancement provided by the originator that is not compatible with the objective of capital relief since this cumulated excess spread will constitute a first loss subject to 1250% RW.
Thus we think that the EBA should allow both features (use-it-or-lose-it and trap).
Question 12: Do you agree with the proposed way to treat the excess spread commitment in synthetic securitisation transactions for the purposes of the quantitative assessment of SRT and commensurate risk transfer?
The FBF members have assumed that this question 12 only relates to a Fixed Excess Spread which would be committed for Synthetic Securitisations. For FBF members’ position on the type of excess spread which should be allowed for Synthetic Securitisations, please refer to answer to question 11.Generally speaking, if the committed Fixed Excess Spread for given period cannot be justified by the actual yield on the portfolio/liabilities costs or is fixed whatever this yield/these liabilities costs, then it should be accounted as additional protection for this period as this is protection in addition to what the portfolio can generate on this period. This additional protection will translate into higher attachment point for the protected tranches – hence the excess spread commitment for a given period will be considered as a first loss retained tranche and likely be subject to 1250% RW/Capital Deduction.
The same would hold for excess spread which would have already been trapped (as this would be equivalent to a cash reserve).
However, the proposal [§123.a III] – which relates to trapped Fixed Excess Spread for a 1-year period - mentions a 1250%RW/Capital Deduction not only for the trapped excess spread but for the future component of the Fixed Excess Spread.
* If the Fixed Excess Spread for the future periods is a fixed percentage of the portfolio outstanding balance of the portfolio at the beginning of each observation period (e.g. 30 bppa applied to the portfolio outstanding balance at the beginning of each quarterly period), it may be difficult to have the amount for the 1-year period accounted as a first loss tranche (presumably subject to 1250%/capital deduction) as the effective amount of Fixed Excess Spread may not be known as the portfolio will amortise.
* If the Fixed Excess Spread for the future periods is simply a fixed amount set in advance of the 1-year period whatever the performance or amortisation of the portfolio (e.g.30bbpa applied to the portfolio outstanding balance at the beginning of each calendar year), this is then equivalent to a non-amortising First Loss retained tranche (which would be presumably subject to 1250%/capital deduction). However, in this case, having such a set-up for this Fixed Excess Spread (equivalent to a fixed first loss) may prevent the capital relief for deals in amortisation phase for example.
Question 13: In relation to the further considerations for stakeholders’ consultation on the own funds treatment of excess spread: a. Do you agree that the unrealised/unfunded component of the excess spread commitment should become subject to Pillar I own funds requirements? b. What would be the impact on SRT transactions if Pillar I own funds requirements were recognised as suggested in Section 3.2?
The FBF members have assumed that this question 13 only relates to a Fixed Excess Spread which would be committed for Synthetic Securitisations as per the §126 and §127.For FBF members’ position on the type of excess spread which should be allowed for Synthetic Securitisations, please refer to answer to question 11. In this answer, the FBF members provide also their position of the treatment of Variable Excess Spread for Synthetic Securitisations – which shall be treated as excess spread for Traditional Securitisations and attract no RW as it is the case for any unsecuritised portfolio (no RW in relation to future margin).
As mentioned in the answer to question 12, there may be different ways of computing the excess spread commitment:
• We assume the case in this answer where it is computed as a fixed amount of the portfolio outstanding balance at the beginning of each observation period (for the other case where the Fixed Excess Spread is a fixed amount set at the beginning of each yearly period whatever the performance and amortisation of the portfolio, please refer to answer to question 12 where the Fixed Excess Spread can be considered as a unprotected first loss tranche).
• In this case, If the portfolio outstanding balance excludes the exposures which have defaulted, this means that the committed excess spread depends on the future performance of the securitised portfolio and hence exists in the future to the extent that the portfolio is performing and has not amortised.
As there is no certainty on this committed Fixed Excess Spread, the FBF members are on the opinion that the unrealised/uncommitted component of Fixed Excess Spread should not attract any RW.
Question 14: Are there any other safeguards or alternative regulatory treatments to address risks retained through excess spread in traditional and synthetic securitisation?
It is absolutely not justified to require recognising excess spread as an additional securitisation position unless it is trapped. The self-assessment should be considered as an adequate and appropriate safeguard to prevent the abuse of the SRT regulations.Question 15: Should there be a specific treatment in those transactions featuring excess spread in which the originator, instead of achieving SRT in accordance with one of the SRT tests specified in the CRR, chooses to deduct all retained securitisation positions from CET 1 or apply a risk weight of 1250% to all of such securitisation positions (‘full deduction option’), in order to be allowed to exclude the securitised exposures from the calculation of risk-weighted exposure amounts?
As the FBF disagrees with the recognition of unrealized excess spread as a securitization position, the FBF also disagrees with recognising it in this situation, since the originator has already effectively reserved sufficient capital to cover even a 100% loss scenario.Question 16: What are your views on the use of originator’s bankruptcy as an early termination clause? How does this clause interact with the resolution regime (i.e. the BBRD framework)? Should this clause be banned?
We are of the opinion that what works best is to make sure that the credit protection remains enforceable as long as the originator does not default on the payment of the premium. In this sense, we are in favor of excluding originator bankruptcy as an early termination event, as long as the originator continues to fulfill its premium or interest payment obligations, all its other reporting and servicing obligations under the transaction subject to the following caveats: this early termination event can only be disapplied if investors are also protected from the counterparty risk caused by depositing the protection tranche funds with the originator (i.e., entirely segregated from originator risk). This can be accomplished by:a. Allowing the investment of cash proceeds in eligible investments held at an independent custodian. According to the latest EBA guidance, this structure would be prohibited for STS SME synthetics and presumably other asset classes if STS benefits are extended to other synthetics.
b. Allowing the cash to be deposited with an originator segregated from the bankruptcy estate of the originator through deposits in a bankruptcy remote custodian account, investments in eligible investments, repos and other similar structures.
c. Downgrade triggers requiring the transfer of deposited funds to another bank or its investment in eligible collateral. The EBA should clarify that such clauses are not problematic for SRT, and that movements of funds due to such clauses would not cause the loss of STS status of a deal.
In some synthetic securitizations of certain asset classes, removing termination in case of originator bankruptcy could have a material impact on rated transactions, as ratings are generally dependent on servicing continuity. This would also be a concern for investors. For example, the performance of larger concentrated assets such as commercial real estate or certain commercial loans, such as SMEs may be highly correlated to the financial health of the originator to the extent servicing staff were made redundant. Accordingly any ban on termination clauses involving originator bankruptcy must take these concerns into account. Investors would have to be given the right to terminate in a bankruptcy situation to the extent that servicing arrangements (and indeed investor reporting obligations) were not carried out satisfactorily.
Question 17: Do you agree with the proposed originator’s self-assessment of risk transfer? Should such assessment be formulated differently?
Preliminary remark- The EBA proposes to use the latest EU wide stress tests parameters as stress scenario on the PD & LGD to the underlying exposures
- We understand that institutions would compare the amount of losses transfered in this scenario of stress to the amount of RWA transfered.
- The amount of RWA represent an amount of unexpected loss which is defined, under the IRB approach, by the RW formula provided by the Basel committee.
- This Unexpected Loss represent a stress scenario on the PD that is much higher than the stress scenario used for the EU stress test :
o For instance, the implicit stress for a corporate with a current PD of 0,10% under the IRB framework is a stressed PD of 3,42% while the EU stress test results in a stressed PD of 0,16%.
o A PD of 1% result in a stressed PD of respectively 14% (IRB) and 1,61% (EU stress test).
- Thus we think that if it relies on the EU stress test the self-assessment of SRT could by construction result in an absence of risk transfer since banks will compare 2 scenarios (IRB vs EU stress test) whose amplitude have no common measure.
Notwithstanding this, with regards to the whole process, our understanding of the EBA’s proposition is that in order to achieve SRT:
- All securitization transactions will need to pass the new Enhanced Quantitative Tests (EQT) - option 1 or 2.
- Synthetic transactions and complex traditional securitizations will need to pass the new enhanced quantitative tests and the new Quantitative Self-Assessment of Risk Transfer (QSART).
- Simple and traditional securitization transactions will need to pass only the new EQT to achieve SRT.
This approach has two shortcomings:
- The quantification of retained and transferred risk in the new EQT relies on regulatory parameters such as EL and own fund requirements which may in some cases be disconnected from the real level of risk of the exposures, especially for banks using the standardised approach and for NPL portfolios.
- The EQT and the QSART have the same purpose. Both intend to assess whether the reduction in own funds requirements is justified by a commensurate transfer of credit risk. Therefore having to pass both tests for synthetic or complex transactions would be redundant.
We propose the following changes to the approach:
- The issuers of simple and traditional securitisation should be allowed to achieve SRT by using the QSART instead of the new EQT. However, the possibility for simple and traditional transactions to achieve SRT by using the EQT should be maintained.
- As the QSART relies on a more accurate measurement of risk than the EQT, transactions which have used the QSART to prove the commensurateness of the risk transfer should not need to pass the EQT. This should be applicable to any transaction regardless of their complexity and their nature (synthetic or traditional).
- These principles should be applicable to NPL transactions as well.
Question 18: Are you aware of circumstances where institutions have entered into a structured risk transfer transaction which is not captured by Articles 243 or 244 CRR? For example, where the accounting treatment has meant a transaction is not considered for SRT assessment, or where transactions economically similar to SRT transactions do not fall into the definition of a ‘traditional securitisation’ or ‘synthetic securitisation’.
No comment.Question 19: Do you agree with the proposed specification of the minimum first loss tranche thickness for the purpose of the first loss test?
We agree that a test for minimum thickness of protection tranches is appropriate. Notwithstanding this, the current mechanistic tests show an anomaly that EBA must correct in order to avoid unjustifiably discarding the achievement of SRT for viable securitization transactions:1. In the first illustration below, 80% of the EL and UL risk is sold to investors, which meets the substantive requirements of the EBA guidance. However, in the first example, the senior retained tranche has been divided into senior and mezzanine tranches (which is the most efficient structure under the SEC-ERBA and SEC-SA approach). Because the mechanistic test states that if mezzanine tranches are present, the originator must sell >50% of the risk weight of the mezzanine tranches, the mechanistic test fails. Nevertheless, in the second example where no mezzanine tranches are present, the first loss test is passed. The first example fails, even though substantively the same amount of risk (which exceeds EL & UL) has been transferred to investors. We believe this is an unintended consequence of the mechanistic tests which should be corrected, if the rest of the EBA guidance is ultimately agreed requiring that a significant portion of the EL and UL risk has been transferred.
[see illustration in the attached file]
2. The minimum first loss test requiring that the first loss tranche cover 100% of EL and 2/3 of UL should be modified to take into account a scenario such as the one illustrated below. In this graph, we see that the originator has transferred 100% of the EL and UL to investors. The first loss tranche does not meet the first loss test, but in substance, the combination of the mezzanine and first loss tranches sold is sufficient to transfer 100% of EL and 2/3 of UL. If not worded correctly, the first loss tranche would inadvertently disallow the illustrated transaction.
[see illustration in the attached file]
3. The revised definition of mezzanine tranches, which excludes mezzanine tranches weighted 1250% exacerbates the problem described above. Under the newly adopted capital calibrations, it is likely that many mezzanine tranches will be weighted 1250% (such as in the case of IRB consumer loans in which the SEC-IRBA approach does not recognise the benefit of excess spread). Accordingly, in the above example, if the most subordinate mezzanine tranche is weighted 1250%, the mezzanine test would fail, even though the originator has transferred risk equal to EL + 2/3 UL.
Question 20: Do you agree with the proposed requirement of the minimum first loss thickness for the transactions assessed under the mezzanine test (i.e. transactions including mezzanine securitisation positions)? Do you consider this requirement relevant for all the approaches for calculation of securitisation own funds requirements (including e.g. SEC-ERBA)?
In principle, we agree with the substance of the test, which we understand is trying to prevent abuses of the test observed by the EBA. However the structure of the test has an unintended negative consequence, as further explained below.The test should be restated or clarified to take into account transactions in which a sufficiently large portion of mezzanine risk is sold to cover UL, and where the attachment point of the sold mezzanine tranches encompasses the portion of EL not covered by a thin first loss tranche. In most cases we find that the optimal cost of protection and RWA reduction can be achieved if a small first loss tranche, is retained, while transferring a sufficient amount of mezzanine tranches to cover the rest of EL and most of UL. To prevent the abuse that this test is intended to cover, the EBA could specify that the mezzanine test will not cause a transaction to fail the minimum required for SRT if the originator transfers any combination of first and mezzanine tranches with a thickness equal to EL + 2/3 UL. Another alternative would be to require banks using the mezzanine test to sell a greater percentage of mezzanine tranches than 50%, such that the percentage sold above 50% equals the amount of EL not covered by the first loss tranche.
Question 21: Is a specification needed of the minimum thickness of tranches constituting mezzanine securitisation positions for the purpose of the mezzanine test?
No specification needed.Question 22: What impact do you expect the new CRR securitisation framework to have on tranches’ minimum thickness?
EBA rules should be consistent with Basel and not impose a minimum size for the first loss tranche.With respect to the new CRR securitization framework, senior risk weights tend to increase dramatically whereas mezzanine risk weights tend to be rather stable and first loss tranches remain to be deducted/with 1250% risk weight. EL and provisioning of the underlying portfolio remain however unchanged.
From the originator’s perspective, the focus is to achieve economically efficient SRT. With senior risk weights being most impacted from the new regulation, originators might tend to structure senior tranches in a way to keep risk weights on retained senior positions as low as possible. This will likely be achieved by higher amounts of subordinated tranche notional (incl. Excess Spread). In this respect, we rather expect senior tranche sizes to shrink and mezzanine as well as first loss tranche sizes (incl. Excess Spread) to increase. However, a more likely effect in most cases is that banks will cease doing SRT deals because it is likely that the incremental cost of selling additional tranches will make the transaction uneconomic compared to the bank’s cost of capital, and the sale of such additional tranches might cause the bank to fail the cost of credit protection test as proposed by the EBA.
From the investor’s perspective, the focus is to take on an investment with an attractive risk/reward profile. Assuming the investor is not regulated under the CRR, the new regulation is not key in the assessment of risks and rewards. The focus is rather on (economic) expected and unexpected losses of the underlying portfolio in order to assess the risk of the investment. We expect the impact of the new CRR regulation on the investor´s minimum tranche thickness requirements to be marginal.
However, those investors most interested in first and second loss tranches have yield requirements higher than the weighted average cost of protection required by originators for a deal to make economic sense. If the above-mentioned anomaly in the first loss and mezzanine tranches is not corrected, it will be difficult, if not impossible, for originators to sell a thick enough portion of tranches at an economic level.
Question 23: Do you have any comments on the test of commensurate risk transfer proposed under Option 1?
In analyzing the impact of these tests on banks existing SRT positions, we note the following problems:1. In many cases the new definition of mezzanine, excluding mezzanine tranches which are risk weighted 1250% causes the deal to fail the mezzanine test, even in deals which have transferred tranches equal to EL+2/3UL. The likelihood of mezzanine tranches being risk weighted 1250% is much greater under the new CRR approaches.
2. As per our responses to questions 19, 20 and 21 above, the minimum first loss tranche thickness test inadvertently causes several of our transactions to fail, even though we have transferred a total amount of tranches exceeding EL+2/3UL.
We would like to mention another inconsistency that particularly seems to occur to synthetic transactions which are structured to pass the mezzanine test and where the originator is obliged to apply SEC-IRBA.
In case the first loss tranche plus excess spread are retained by the originator we acknowledge the introduction of the minimum first loss piece thickness test as a helpful measure to ensure that the lifetime EL of a portfolio is covered by the originator whereas its UL is transferred. Given this intention, it is unclear why the lifetime EL is considered in ratio 2 of the commensurate risk transfer test. In fact, sections 216 and 218 reveal that ratio 2 of aims at measuring “the risk transferred to the third parties” obtained “as the share of losses transferred to third parties for the life of the transaction”. We understand that the commensurateness test is not solely directed towards transactions that intend to pass the mezzanine SRT test. However, it is our understanding that the test of commensurate risk transfer contradicts the minimum tranche thickness test particularly in case the mezzanine SRT test applies. This is due to the fact that the latter requires the lifetime EL not to be part of the transferred risk whereas the commensurateness test considers it to be transferred.
Furthermore, the reason why the commensurate risk transfer test compares a 1y measure, ratio 1, with a lifetime measure, ratio 2 is not clear. Even though it is clear why comparing capital savings to transferred risk, it remains questionable why actually basing these compared figures on different time horizons.
Also it seems that Ratio 2 takes into account the lifetime excess spread: if the excess spread is committed and trapped, indeed this should be considered as a retained position – however as detailed in earlier questions, unearned excess spread or un-trapped excess spread should not be considered as a retained position
Question 24: Do you have any comments on the test of SRT and commensurate risk transfer proposed under Option 2? In particular, is the 50% threshold for SRT therein needed and appropriate?
See comments in question 23 above. In principle, if we accepts this type of test, it should not matter whether the originator sells 20% of the risk or 100% of the risk, since the RWA reduction achieved will be commensurate with the risk transferred.However, it is important to note that for synthetic securitizations where synthetic excess spread of the amount of the EL is trapped on an annual basis, the 1st Excess Spread payment that is trapped towards the end of the 1st year should not be considered in the calculation of Condition 1 since it serves as a buffer of the EL of the underlying portfolio that materializes during the 2nd year of the transaction.
In analysing the impact of this test on banks existing SRT positions, we note that, in many cases, the increased capital on senior retained tranches required by the new capital calibrations under the revised CRR causes our SRT transactions to fail, even though the positions sold to investors exceed EL+2/3UL, which is the minimum suggested by the EBA. The effect of the new calibrations is to require an originator to sell a much larger portion of first loss and/or mezzanine tranches in order to reduce the risk weight of the retained senior tranches below the threshold required by the test. For any particular transaction, some or all of the following negative impacts then may apply:
a. It increases the cost of protection above the bank’s cost of capital, making the deal uneconomic to execute.
b. It causes the total cost of protection to exceed the EBA guidelines on the cost of credit protection.
c. It requires the originator to sell total tranche thicknesses well in excess of EL+2/3UL, which makes such test redundant or inappropriate.
Question 25: Should the SRT test be different depending on asset classes? Should it differ across STS and non-STS transactions?
No differentiation based on asset classes is necessary. Instead, there should be a differentiation in the test according to the capital approach that an originator is required/chooses to use, in order to take into account the double penalty of capital inherent to the SEC-IRBA and SEC-SA approaches, especially for non-auto transactions with high excess spread portfolios, which distort the capital charge of the tranches, hence the need to sell a higher number or portion of tranches than is necessary to cover EL and UL.Question 26: Could you provide, on the basis of SRT transactions that are part of your securitisation business, an assessment of the impact in terms of SRT achievement of the proposed requirements under both Option 1 and Option 2, taking into account the new CRR securitisation framework (Securitisation Regulation package)?
No commentQuestion 27: Do you agree with the assessment of the market practice of NPL transfer? Are there material aspects that are not covered in this representation?
We agree with the general overview of the market of the NPL transfer presented in the Discussion Paper. However, we would like to point out the following aspects:• Generally, the key issue regarding NPL transfer is that the NPL are not anymore included on the balance sheet of the originator and hence it obtains a decrease of its NPL rate.
• The SRT and regulatory capital relief are usually not key drivers in our sense for such NPL transfers.
• Consequently, the question of who will be in charge of the servicing, the originator or an independent third party manager, is critical for the structuring of NPL securitizations. As a matter of fact, when the originator of the assets remains the manager/servicer after the assets have been “transferred” to the SSPE, the assets can usually not be derecognized or the SSPE be deconsolidated in application of the IFRS accounting rules (control test).
• Another point of attention is the diverging views that the originator and the investors can have on the expected cash flows.
Question 28: What conditions/initiatives would, in your view, facilitate the well-functioning of the NPL securitisation market?
The NPL secondary market could be further developed if certain obstacles are removed:• Barriers to entry for NPL investors: in accordance with French regulation, loan servicers have to hold a licence to acquire and service French loans. It remains a hurdle for the development of the NPL market. Moreover, to acquire large NPL portfolios, buyers will have to acquire and potentially develop existing dedicated platforms that have sufficient trained and knowledgeable staff and the technical capacity to service large numbers of loans.
• Limitations on the transferability of loans: banks wishing to sell any loan portfolios are required to remove the loans which include a clause of non-transferability into the underlying contract. Such a clause continues to be in force once a loan becomes non-performing.
• Disclosure limitations during certain legal proceedings: banks are not allowed to disclose to potential buyers that a mandate or conciliation procedure is underway, which complicates the sale of any NPLs affected by such a legal proceeding.
In addition, public sector/state guarantee on NPL securitisations available in other countries like Italy with efficient payment mechanisms could facilitate the distribution / placement of the NPL securitisation transactions. This guarantee could cover the senior notes and/or as the case may be, the mezzanine notes.
Question 29: Which, in your view, are the core structural features that should be assessed within the SRT assessment of NPL securitisation transactions? Are the proposals on selected structural features of securitisation transactions proposed in this document (see Section 3.2.2) equally valid for NPL securitisation transactions?
As the NPL transaction are usually/ exclusively cash traditional securitizations without revolving period, the proposed structural features for traditional securitization are appropriate except the excess spread which is not relevant for such asset class.Nevertheless, we think that the future EBA guidelines will need to take into account future regulation from the EC and the guidance from the ECB on the prudential provisioning of NPL. Indeed the application of the proposal made by the EBA under figure 25 of the discussion paper to a NPL portfolio would result in significant capital relief if the future securitization framework does not take in account the prudential provisioning and thus could lead to significant incentive for securitization and regulatory arbitrage.
Question 30: Do you agree with the proposed way of implementing the SEC-IRBA and SEC-SA approaches for the calculation of securitisation tranche capital in the presence of a non-refundable purchase price discount? Do you envisage other ways to implement the mentioned approaches in the presence of a non-refundable purchase price discount?
We agree with the way of implementing the SEC-IRBA. We note that this method should be generalized to the securitisation transactions where there is a non-refundable purchase price discount irrespective of the type of underlying assets (i.e. not just used for NPL securitization).We also note that there is a wide discrepancy between the results of the SEC-IRBA and the SEC-SA on the Risk Weights of the different classes of notes of the NPL securitisation example (Fig. 25).
In addition, the RWA of the junior notes seems at a first glance relatively low compare to the RWA of other mezzanine or junior notes of usual securitisations. On the contrary, the RW on all the notes with the SEC SA are extremely high and in particular on the senior tranche.
The overly conservatism of the SEC-SA for NPL pools stems in our view from the way the ‘W’ factor is taken into account in the SEC-SA. This issue is not specific to NPL but it is amplified in the case of NPL pools. It was already identified with a simple correction proposed in a paper called: “How to revive the European Securitisation Market: a Proposal for a European SSFA” [*] by Georges Duponcheele Alexandre Linden and William Perraudin published in November 2014. Indeed the paper mentions: “It is to be noted that this official version of the SA in the BCBS 269 proposals contains a conceptual error in that it includes the delinquent assets (via Ka) in Kp. This in effect means that the capital associated with provisions for the delinquent assets is allocated to the senior tranches via the exponential function and the parameter a.
To correct this conceptual error, Kp should simply equal pool capital for performing loans and should not be a function of W. The delinquent assets should only affect the 100% capital charge threshold KT. It is important to distinguish between KT and KP to avoid making errors.”
We therefore reiterate the need to correct the way the W factor is taken into account in the SEC-SA as detailed in the extract above.
[*] https://www.researchgate.net/publication/279960182_How_to_Revive_the_European_Securitisation_Market_a_Proposal_for_a_European_SSFA_Alexandre_Linden_William_Perraudin_BNP_Paribas_BNP_Paribas_Risk_Control_Limited
Question 31: Do the SRT quantitative tests provided for in the CRR currently in force (Articles 243 and 244 of the CRR) work properly for NPL securitisation transactions? If not, please provide an explanation to your answer.
We think that clarification is required from EBA on how the non-refundable purchase price would be taken into account for SRT quantitative test purposes. This would be necessary not just for NPL but also for other cases where there is a non-refundable purchase price discount.Moreover, SRT is not the main driver for NPL securitizations; major drivers are the search for senior refinancing or decrease of the balance sheet NPL rate (see our answer to question 27)