Response to consultation on changes to ITS on reporting
Go back
Member firms analysed that the EBA’s intention, through this measure is to require institutions to report the relative positioning of the ‘Fair Value’ and the ‘Expected Value’, where the ‘Expected Value’ is the one referred to in the appendix of the RTS.
At a valuation exposure level, and with the idealized case where a distribution is available using quantitative ranges, the knowledge of the expected value, fair value and of the 90% confidence prudent value, provides the sufficient arguments to understand the AVAs, for a given choice of aggregation methodology. However, when taken at aggregate level, such simplistic approach fails providing relevant information given that the RTS doesn’t allow compensation between prudence excess and prudence deficit.
We therefore fear that the interpretation, at an aggregate level, will not allow supervisors to derive valid conclusions about the AVAs.
In addition to the above, we observe that the definition of an upside value, would require a full set of guidance to be developed within the RTS in a similar way than the downside, to avoid confusion. Such confusion may arise from the fact that fair value adjustments may in some instances stand higher than both the 10% and 90% percentiles (example of AVA categories that are not allowed under IFRS13), and in some others stand lower than both, and finally can also stand within the range (likely case for Close out Cost). Furthermore, it is unclear how diversification enters into play in the circumstance.
Ultimately, we expect that the cost of implementation of an upside uncertainty measure might be disproportionate to the likely usefulness of the aggregate measure required to be reported in the template
As a result, and as explained in the previous Question, it is difficult to believe that including a well-defined upside uncertainty would pass the cost-benefit analysis.
Finally, member firms have not prepared themselves at all for such a new reporting requirement and as such, would definitely require a significant amount of additional work that cannot be expected to be finalised within a short time frame.
We are concerned however with the requirement to provide a breakdown of assets and liabilities by portfolios as such information is not easy to source from the accounting ledgers in many institutions. Given that the primary purpose is to provide a split amongst those valuation adjustments using the core approach and those who are based on the fall back method, we propose that the template to solely require the reporting of fair value assets and liabilities at the level of trading book/banking book only. This means that only cells {130x010; 140x010; 130x020; 140x020; 130x040; 140x040; 130x170; 140x170; 130x260;140x260 } of the template C32.02 are used.
As far as the fair value and prudent value adjustments are required, the concern relates to the prescriptive nature of the portfolio breakdown in column C of the template C32.02. While we understand that the EBA aims at providing some basis of comparability in the reporting, and while we appreciate that this template is similar to the one used during the QIS and to the one used in the UK prudent valuation return, we believe that this imposed reporting will fail the use test. Member firms considered that as long as the reported figures do not correspond to an existing risk management axis, hardly would such template be useful in the context of internal monitoring and reporting of the prudent valuation amounts.
We suggest that the EBA contemplates leaving the portfolio split to be decided by each institutions using a simplified, yet existing, business hierarchy, and also leaving institutions to decide which of the items in the EBA hierarchy is the closest description.
Member firms are also concerned with the requirement to report QTD Revenue for portfolios (in Column 150). They expressed that they would not consider the revenue attributed to portfolios to which the fair value positions were assigned to be useful in meeting the objectives described in the explanatory note preceding question 3. As a matter of fact, the EBA did not make a case to explain why this was required or is deemed relevant for the purpose of understanding the magnitude of valuation uncertainty or AVAs. Member firms would therefore welcome the removal of this requirement from the template.
As touched upon in Question 3, portfolios are organised based on business organisations rather than on the degree of complexity of products. Complying with a strict interpretation of the template would be extremely difficult and it is not clear that it would significantly enhance the understanding of the Prudent Valuation calculations.
Again, member firms expressed preference to an alternative consisting in using existing business hierarchies as a starting point, and then distributing (mapping) them on a best effort basis, according to the proposed ITS classification. This can be based on the main risk drivers or other macro measure and would be specific to each institution rather than imposed through a reference to regulation.
Member firms indicated that even with such simplification, significant additional work would remain in order to calculate AVA on such a granular basis, whereas they observe that the interpretation of line 180 of template C32.02 may prove complex due to likely presence of netting across portfolios.
In turn, they indicated, based on their Asset Quality experiences, that there is possibly case to slightly ease the burden of reporting through restricting the number of Model / Product to 20. Member firms expect this will still ensure 95% coverage of the Model AVA. A similar limitation could also be applied to the Concentration template (C32.04); yet other concerns were expressed in relation to this template.
Membership assessed that EBA should give due consideration, when deciding the timing of implementation, that these information are likely to require significant workload, and are very likely to require industrial capabilities to be invested in as the number of calculations have been de facto multiplied by a factor higher than 20.
The other issue in requiring such information is the EBA ITS may not have provided sufficient guidance on these concepts, to ensure that the investment will pass the the cost benefit ratio, especially that none of these extra information has any effect on the final AVA.
Ultimately, memberships expressed preference to the simplification (business hierarchy) or removal (upside uncertainty and overhedges) of these new requirements rather than seeking for additional guidance or an extra/phased implementation time. This is essentially because they did not believe that their inclusion would pass the cost benefit ratio.
As we have already discussed extensively the question of business hierarchies and “Upside Uncertainty” above, we will solely discuss the question of overhedges in the below.
“Overhedges” is defined as: Where the valuation model is deliberately set up to model a set of terms that differ from the legal terms of a position (e.g. for the purposes of achieving a more manageable risk profile) the aggregate effect such differences on the valuation of the relevant sub-portfolio shall be reported in this column. Examples include barrier shifts and representation of a digital option as a put - call spread.
First it must be highlighted that such calculations are not necessarily performed on an on-going basis by institutions. As a result development would be required to comply with such reporting.
Secondly, and as the EBA text acknowledges it, the alternative representation of the legal payoff is usually a reflexion that the replication strategy implied by the legal features cannot be implemented, and that therefore alterations are needed as a starting point for pricing, hedging and valuation uncertainty measurement. This practice is largely shared by market participants and is reflected in any exit price or risk measure.
If the EBA was to consider or suspect that overhedge need to be covered through any specific RTS requirement (uncertainty, control, governance etc…), we believe that specific guidance should be developed at the first place, possibly following the normal policy development cycle (including a QIS if needed). If not, we believe that its inclusion in the template is costly and possibly useless.
It is however our view that it would not be appropriate to impose this field for the sole purpose of providing information without explaining the purpose pursued.
In addition, we would like to highlight that the structure of the concentrated position template mixes the product axis and the risk axis in a confusing manner.
We believe that the natural axis for this template should be the valuation exposure given that the concentrated position AVA is an incremental cost to the Close out Cost. In consistency with this, the columns H (060) and I (070) should be withdrawn and the columns 010, 020, 030 solely required to provide the main Asset class/product/underlying that composes the valuation exposure.
Also, consistently with our comment on model risk, we propose this template to be limited to 20 largest exposures to ease reporting burden.
Question 1. The EBA believes that an understanding of where the accounting fair value sits within the notional range of plausible values at an aggregate level is essential context for assessing that the downside of this range, and therefore the AVA, is appropriately reported. Do you agree with this statement? If not please explain your reasoning? [Annex 2, page 1]
Member firms disagreed with this statement, and expressed strong concerns about the introduction of a hard-to-obtain measure, that is neither required by the RTS, nor meant to have any effect on the AVA calculations. They also failed understanding how the positioning of the fair value within the notional range of plausible values can provide a basis to be confident about appropriateness of the reported AVAs.Member firms analysed that the EBA’s intention, through this measure is to require institutions to report the relative positioning of the ‘Fair Value’ and the ‘Expected Value’, where the ‘Expected Value’ is the one referred to in the appendix of the RTS.
At a valuation exposure level, and with the idealized case where a distribution is available using quantitative ranges, the knowledge of the expected value, fair value and of the 90% confidence prudent value, provides the sufficient arguments to understand the AVAs, for a given choice of aggregation methodology. However, when taken at aggregate level, such simplistic approach fails providing relevant information given that the RTS doesn’t allow compensation between prudence excess and prudence deficit.
We therefore fear that the interpretation, at an aggregate level, will not allow supervisors to derive valid conclusions about the AVAs.
In addition to the above, we observe that the definition of an upside value, would require a full set of guidance to be developed within the RTS in a similar way than the downside, to avoid confusion. Such confusion may arise from the fact that fair value adjustments may in some instances stand higher than both the 10% and 90% percentiles (example of AVA categories that are not allowed under IFRS13), and in some others stand lower than both, and finally can also stand within the range (likely case for Close out Cost). Furthermore, it is unclear how diversification enters into play in the circumstance.
Ultimately, we expect that the cost of implementation of an upside uncertainty measure might be disproportionate to the likely usefulness of the aggregate measure required to be reported in the template
Question 2: Would the ‘upside uncertainty’ measure defined above and used in column 120 be suitable as a definition of the upside uncertainty? If not please provide reasons and any alternative suggestions for how such an upside measure could be defined. [Annex 2, page 1]
Member firms expressed consensus that calculating a 10% percentile will largely require doubling the workload in most of the non-trivial cases that naturally are the most resource consuming.As a result, and as explained in the previous Question, it is difficult to believe that including a well-defined upside uncertainty would pass the cost-benefit analysis.
Finally, member firms have not prepared themselves at all for such a new reporting requirement and as such, would definitely require a significant amount of additional work that cannot be expected to be finalised within a short time frame.
Question 3: Is the above approach to splitting out fair valued assets and liabilities and fair-value adjustments on the one hand between the different types of AVAs and on the other hand between asset classes and product categories practical to implement? If not please describe the practical obstacles. Please suggest any alternative approaches (particularly if an alternative approach has been found useful for internal reporting purposes). [Annex 2, page 5]
We support, in principle, the consultation’s proposal to disclose fair value adjustments using the same categories than their Prudent Value equivalent. Although such split is not required by the accounting standards, it is likely to have been already performed by member firms for the purpose of aggregating the AVAs; therefore we do not expect significant cost to arise from such requirement.We are concerned however with the requirement to provide a breakdown of assets and liabilities by portfolios as such information is not easy to source from the accounting ledgers in many institutions. Given that the primary purpose is to provide a split amongst those valuation adjustments using the core approach and those who are based on the fall back method, we propose that the template to solely require the reporting of fair value assets and liabilities at the level of trading book/banking book only. This means that only cells {130x010; 140x010; 130x020; 140x020; 130x040; 140x040; 130x170; 140x170; 130x260;140x260 } of the template C32.02 are used.
As far as the fair value and prudent value adjustments are required, the concern relates to the prescriptive nature of the portfolio breakdown in column C of the template C32.02. While we understand that the EBA aims at providing some basis of comparability in the reporting, and while we appreciate that this template is similar to the one used during the QIS and to the one used in the UK prudent valuation return, we believe that this imposed reporting will fail the use test. Member firms considered that as long as the reported figures do not correspond to an existing risk management axis, hardly would such template be useful in the context of internal monitoring and reporting of the prudent valuation amounts.
We suggest that the EBA contemplates leaving the portfolio split to be decided by each institutions using a simplified, yet existing, business hierarchy, and also leaving institutions to decide which of the items in the EBA hierarchy is the closest description.
Member firms are also concerned with the requirement to report QTD Revenue for portfolios (in Column 150). They expressed that they would not consider the revenue attributed to portfolios to which the fair value positions were assigned to be useful in meeting the objectives described in the explanatory note preceding question 3. As a matter of fact, the EBA did not make a case to explain why this was required or is deemed relevant for the purpose of understanding the magnitude of valuation uncertainty or AVAs. Member firms would therefore welcome the removal of this requirement from the template.
Question 4: Is the above portfolio-based approach to splitting out AVAs and other attributes between ‘Exotic’ and ‘Vanilla’ practical to implement? If not please describe the practical obstacles. Please suggest any alternative approaches (particularly if an alternative approach has been found useful for internal reporting purposes). [Annex 2, page 12]
Most members calculate AVA based on valuation inputs on an institution-wide basis and not on a portfolio or group of portfolio basis. In general, AVA will not be attached to a portfolio and for example “exotic” portfolio will almost always generate AVAs coming from “vanilla” risk factors.As touched upon in Question 3, portfolios are organised based on business organisations rather than on the degree of complexity of products. Complying with a strict interpretation of the template would be extremely difficult and it is not clear that it would significantly enhance the understanding of the Prudent Valuation calculations.
Again, member firms expressed preference to an alternative consisting in using existing business hierarchies as a starting point, and then distributing (mapping) them on a best effort basis, according to the proposed ITS classification. This can be based on the main risk drivers or other macro measure and would be specific to each institution rather than imposed through a reference to regulation.
Member firms indicated that even with such simplification, significant additional work would remain in order to calculate AVA on such a granular basis, whereas they observe that the interpretation of line 180 of template C32.02 may prove complex due to likely presence of netting across portfolios.
Question 5: Do you think such mismatches between the portfolio-level AVAs and the institution-level AVAs would be significant? Please give examples. [Annex 2, page 12]
Some risk factors like FX and IR are definitely cross product line. Firm members agreed that it is reasonable to expect exposures to such risk factors may be limited in portfolios covering other asset classes. Nevertheless they agreed that significant offsetting can still remain even when the overall exposure is limited.Question 6: Where the difference is significant what additional practical difficulties would arise from calculating AVAs for each of the portfolio categories in rows 050-170? [Annex 2, page 13]
When the calculation process is built around institution level aggregation of valuation exposures, any more granular reporting will require a review of the current process that will be time consuming for some members.Question 7: What are stakeholders’ views on the ability to usefully summarise in a few key words the models and products concerned, as well as on the associated reporting burden or IT issues? [Annex 2, page 15]
Member firms see no specific issue with the requirement to provide textual description.In turn, they indicated, based on their Asset Quality experiences, that there is possibly case to slightly ease the burden of reporting through restricting the number of Model / Product to 20. Member firms expect this will still ensure 95% coverage of the Model AVA. A similar limitation could also be applied to the Concentration template (C32.04); yet other concerns were expressed in relation to this template.
Question 8: Do you find the proposed instructions on prudent valuation clear? Are there specific parts where definitions or instructions should be clarified?
As explained before, the ITS contain new information that was not expected to be required in the RTS (upside uncertainty, granular AVA calculations, overhedges).Membership assessed that EBA should give due consideration, when deciding the timing of implementation, that these information are likely to require significant workload, and are very likely to require industrial capabilities to be invested in as the number of calculations have been de facto multiplied by a factor higher than 20.
The other issue in requiring such information is the EBA ITS may not have provided sufficient guidance on these concepts, to ensure that the investment will pass the the cost benefit ratio, especially that none of these extra information has any effect on the final AVA.
Ultimately, memberships expressed preference to the simplification (business hierarchy) or removal (upside uncertainty and overhedges) of these new requirements rather than seeking for additional guidance or an extra/phased implementation time. This is essentially because they did not believe that their inclusion would pass the cost benefit ratio.
As we have already discussed extensively the question of business hierarchies and “Upside Uncertainty” above, we will solely discuss the question of overhedges in the below.
“Overhedges” is defined as: Where the valuation model is deliberately set up to model a set of terms that differ from the legal terms of a position (e.g. for the purposes of achieving a more manageable risk profile) the aggregate effect such differences on the valuation of the relevant sub-portfolio shall be reported in this column. Examples include barrier shifts and representation of a digital option as a put - call spread.
First it must be highlighted that such calculations are not necessarily performed on an on-going basis by institutions. As a result development would be required to comply with such reporting.
Secondly, and as the EBA text acknowledges it, the alternative representation of the legal payoff is usually a reflexion that the replication strategy implied by the legal features cannot be implemented, and that therefore alterations are needed as a starting point for pricing, hedging and valuation uncertainty measurement. This practice is largely shared by market participants and is reflected in any exit price or risk measure.
If the EBA was to consider or suspect that overhedge need to be covered through any specific RTS requirement (uncertainty, control, governance etc…), we believe that specific guidance should be developed at the first place, possibly following the normal policy development cycle (including a QIS if needed). If not, we believe that its inclusion in the template is costly and possibly useless.
It is however our view that it would not be appropriate to impose this field for the sole purpose of providing information without explaining the purpose pursued.
Question 9: Do respondents have any comments on the structure and content of the proposed templates on prudent valuation?
Most of the important comments have been expressed in the previous questions.In addition, we would like to highlight that the structure of the concentrated position template mixes the product axis and the risk axis in a confusing manner.
We believe that the natural axis for this template should be the valuation exposure given that the concentrated position AVA is an incremental cost to the Close out Cost. In consistency with this, the columns H (060) and I (070) should be withdrawn and the columns 010, 020, 030 solely required to provide the main Asset class/product/underlying that composes the valuation exposure.
Also, consistently with our comment on model risk, we propose this template to be limited to 20 largest exposures to ease reporting burden.