Response to consultation on draft Regulatory Technical Standards (RTS) on prudent valuation
Go back
• Related to Article 3, paragraph 2 we like to comment that Settlement prices of clearing should also be part of the range of possible valuation inputs.
• We like to add that the simplified approach will normally be applied by smaller banks. Certain smaller banks make up their balance sheet under local GAAP. Depending on the local GAAP and chosen valuation principles, significant deviation can therefore occur between banks that classify financial instruments at amortized costs or on fair value. Therefore having the same risk from a market and credit risk point of view could still end up with significant different prudent valuation adjustments.
• Related to Article 8, 4b (Calculation of market price uncertainty AVA ; volatility ratio), it is not fully clear how these volatility measures could be measured. The normal P&L volatility would be clear as it is just the daily P&L based on full revaluation but calculating the P&L on the reduced input parameters cannot directly be done. The normal P&L would require inputs for the complete curve. These can be done on the basis of the reduced inputs. Normally building such a curve would in essence already be based on interpolation between the liquid points. No volatility should arise out of such a calculation. Normal P&L would also be subject to other elements (simultaneous changing of risk factors, realised P&L). One should therefore create Hypo P&L based on multiple ceteris paribus curves. In the full revaluation scheme this would create too much a daily burden. An alternative would be that the daily changes in rates multiplied by the granular deltas (or other exposures) are compared with the reduced delta position time the daily change in the reduced parameter.
• Although it is appreciated that examples are given to explain the consultation paper, these are stylised oversimplified examples. We understand from EBA that these examples are for reference only and that EBA prefers a principle based approach.
Unearned credit spreads AVA
• With respect to Article 10, par 1a we have a remark(“An AVA for market price uncertainty, as defined in Article 8, for all valuation exposures relevant to the calculation of ACVA”):
o With regard to CVA already a prudent valuation adjustment is taken by not taking into account the DVA. If we would have a prudent assessment of fair value one should look at the combination of CVA and DVA and make a prudent assessment of the combined effect. As the DVA is disregarded adding adjustments for the CVA would be stacking of adjustments.
• We do not agree on investing and funding cost as this topic is still under discussion. As IAS 39 applies a mixed valuation model approach (certain positions at fair value while others at cost), funding cost is no- existent in case positions at fair value are funded by financial instruments valued at fair value, or a timing matter in case they are funded by financial instruments valued at amortized cost. Applying for a investing and funding AVA is therefore not crystal clear and could lead to double counting.
• If positions are fair valued (prudentially) taking into account valuation adjustments as defined in IFRS (including DVA on derivatives) and liquidity /concentration one should assume that the positions can be unwound and that future funding costs are not relevant for current prudential valuation.
Operational risk AVA
• We do not agree on the AVA for operational risk as this is already covered in the operational risk prudential reporting. Furthermore we think that the 10% sum of the market price uncertainty AVA’s is rather high as banks that do not apply the AMA method already, calculate the operational risk prudently. We rather think that a 2% sum of the market price uncertainty AVA’s is appropriate,
Model Risk AVA
• Although, we understand the approach of EBA on determining the model risk AVA, from a practical point of view we like to note that for specific models there are no alternatives.
Future Administration Costs AVA
• In case the EBA keeps this as a general requirement for all positions we would advocate some more specific rules in order to ensure a level playing field and comparability.
Early termination AVA
• With respect to Article 15- Calculation of Early termination AVA. We think that it seldom happens that institutions lose on early terminations given the legal requirements and documentation. The rule would require the bank to implement an administration for hardly occurring events. The requirement would entail keeping track of a number of otherwise irrelevant data as number of client trades per types of derivatives. Materiality here isn’t a factor. We would propose that bank keep track of unwinding’s above a certain threshold and on the basis of these numbers make an expert judgement whether this warrants a prudential valuation.
• Investment and Funding AVA. If positions are fair valued (prudentially) taking into account valuation adjustments as defined in IFRS (including DVA on derivatives) and liquidity /concentration one should assume that the positions can be unwound and that future funding costs are not relevant for current prudential valuation.
• An alternative approach could be to state that banks needs to evaluate yearly the liquidity of their positions and substantiate why they consider certain positions liquid due to independent ‘bona fide’ quotes. In addition it would be better that banks are required to document how they have substantiated the fair value of less liquid positions (i.e. what information and judgment was used) that is reviewed periodically by supervisors and auditors. This will enhance the internal controls within banks significantly.
• Additionally, from a practical point of view ongoing monitoring will not be implementable and will require a significant amount of IT investments and human resources that will go beyond the ‘theoretical’ advances that EBA proposes.
• With respect to Article 20 . In particular paragraph 2 is not entirely clear on what is required. A worked out example would be welcomed. As far as the requirement is clear we don’t deem it realistic. Prudential valuation assessment cannot be a daily exercise. As such comparing transaction prices which will be based on daily inputs to prudential valuation inputs maybe meaningless. The as for prudential adjustment corrected contractual prices would be based on numerous prudential valuation inputs. As such comparing contractual prices would not signify much, besides that a prudent valuation would always be less than the contractual prices. It would also create a considerable operational burden. There should be some thresholds in place and restrictions so that for example plain vanilla transactions (FX forwards) are excluded.
Do you agree with the minimum list of alternative methods and sources of information defined above for expert based approaches? If not, what others could be included, or which points from the current list should be removed? State your reasons.
• In our view they seem reasonable and are written broadly enough for banks to implement them in their prudent valuation policy. However the expression “alternative methods and sources of information shall be considered, including all of the following” appears to stipulate they all must be used all of the time. We feel that “all of” should be deleted from this sentence as not all items will be applicable in all situations. We don’t agree that they all have to be considered in each case.• Related to Article 3, paragraph 2 we like to comment that Settlement prices of clearing should also be part of the range of possible valuation inputs.
Do you agree with the proposed simplified approach? Do you think the risk sensitiveness of the approach is appropriate? Are there alternative approaches that you believe would be more appropriate? State your reasons.
• When applying the simplified approach for securities positions the application of FIFO, LIFO or average purchase price can have a significant impact on the amount of unrealised profit and therefore on the AVA. Therefore, the valuation method of inventory items requires further guidance. It should also be noted that most IT applications do not record the unrealized profit of an individual position since it has come on the balance sheet as accounting standards normally do not require to post the unrealized result or only requires this information on an accounting category level (i.e. held to maturity, available for sale, ...) year to date.• We like to add that the simplified approach will normally be applied by smaller banks. Certain smaller banks make up their balance sheet under local GAAP. Depending on the local GAAP and chosen valuation principles, significant deviation can therefore occur between banks that classify financial instruments at amortized costs or on fair value. Therefore having the same risk from a market and credit risk point of view could still end up with significant different prudent valuation adjustments.
Could a differentiated treatment for some asset/liability classes be considered, for example with regard to their liquidity? Please state the pros and cons of such a differentiation. How would you define the degree of liquidity of an asset/liability class (e.g. fair value hierarchy, eligibility for the LCR, other)?
• It would perhaps be more appropriate and more simple to just require for all assets measured at fair value to subtract a fixed percentage, except for highly liquid positions. Highly liquid should therefore be specifically described where a reference to positions for which a two way market exist (as used to exclude positions in the core approach) or to refer to the LCR classification as this is clearly defined and the information is already available.Do you agree with the approach defined above to calculate an AVA where the approaches in Article 8 and 9 are not possible for a valuation exposure? If not, what other approach could be prescribed? Explain your reasoning.
• We consider the 100% of the unrealized profit and either 10% of the notional value or 25% of the market value extremely conservative. The question we raise is what EBA means by ‘not possible’. The case that an institution does not have a clue is mostly the case in distressed markets where liquidity is nil. This measure could result in a self fulfilling property that when markets are closed the capital charge for banks will increase significantly. We assume that this requires more study by academics.Do you agree with the approaches defined above to calculate AVAs for market price uncertainty, close-out costs, and unearned credit spreads? If not, what other approach could be prescribed? State your reasons.
Market price uncertainty AVA• Related to Article 8, 4b (Calculation of market price uncertainty AVA ; volatility ratio), it is not fully clear how these volatility measures could be measured. The normal P&L volatility would be clear as it is just the daily P&L based on full revaluation but calculating the P&L on the reduced input parameters cannot directly be done. The normal P&L would require inputs for the complete curve. These can be done on the basis of the reduced inputs. Normally building such a curve would in essence already be based on interpolation between the liquid points. No volatility should arise out of such a calculation. Normal P&L would also be subject to other elements (simultaneous changing of risk factors, realised P&L). One should therefore create Hypo P&L based on multiple ceteris paribus curves. In the full revaluation scheme this would create too much a daily burden. An alternative would be that the daily changes in rates multiplied by the granular deltas (or other exposures) are compared with the reduced delta position time the daily change in the reduced parameter.
• Although it is appreciated that examples are given to explain the consultation paper, these are stylised oversimplified examples. We understand from EBA that these examples are for reference only and that EBA prefers a principle based approach.
Unearned credit spreads AVA
• With respect to Article 10, par 1a we have a remark(“An AVA for market price uncertainty, as defined in Article 8, for all valuation exposures relevant to the calculation of ACVA”):
o With regard to CVA already a prudent valuation adjustment is taken by not taking into account the DVA. If we would have a prudent assessment of fair value one should look at the combination of CVA and DVA and make a prudent assessment of the combined effect. As the DVA is disregarded adding adjustments for the CVA would be stacking of adjustments.
Do you agree with the approaches defined in Articles 11 to 16 to calculate the various categories of AVAs? If not, what other approach could be prescribed for each AVA? State your reasons.
Investing and funding costs AVA• We do not agree on investing and funding cost as this topic is still under discussion. As IAS 39 applies a mixed valuation model approach (certain positions at fair value while others at cost), funding cost is no- existent in case positions at fair value are funded by financial instruments valued at fair value, or a timing matter in case they are funded by financial instruments valued at amortized cost. Applying for a investing and funding AVA is therefore not crystal clear and could lead to double counting.
• If positions are fair valued (prudentially) taking into account valuation adjustments as defined in IFRS (including DVA on derivatives) and liquidity /concentration one should assume that the positions can be unwound and that future funding costs are not relevant for current prudential valuation.
Operational risk AVA
• We do not agree on the AVA for operational risk as this is already covered in the operational risk prudential reporting. Furthermore we think that the 10% sum of the market price uncertainty AVA’s is rather high as banks that do not apply the AMA method already, calculate the operational risk prudently. We rather think that a 2% sum of the market price uncertainty AVA’s is appropriate,
Model Risk AVA
• Although, we understand the approach of EBA on determining the model risk AVA, from a practical point of view we like to note that for specific models there are no alternatives.
Future Administration Costs AVA
• In case the EBA keeps this as a general requirement for all positions we would advocate some more specific rules in order to ensure a level playing field and comparability.
Early termination AVA
• With respect to Article 15- Calculation of Early termination AVA. We think that it seldom happens that institutions lose on early terminations given the legal requirements and documentation. The rule would require the bank to implement an administration for hardly occurring events. The requirement would entail keeping track of a number of otherwise irrelevant data as number of client trades per types of derivatives. Materiality here isn’t a factor. We would propose that bank keep track of unwinding’s above a certain threshold and on the basis of these numbers make an expert judgement whether this warrants a prudential valuation.
Are there cases where the above AVAs may have a zero value that could be defined in the RTS? If yes, please specify.
• Early termination AVA. Positions will be unwound at market value and an unwinding fee, making it very unlikely that a loss will be incurred due to early terminations.• Investment and Funding AVA. If positions are fair valued (prudentially) taking into account valuation adjustments as defined in IFRS (including DVA on derivatives) and liquidity /concentration one should assume that the positions can be unwound and that future funding costs are not relevant for current prudential valuation.
Do you agree with the approach defined above for the aggregation of valuation exposure level AVAs within the market price uncertainty and close-out cost AVA categories? If not, what other approach could be prescribed? State your reasons.
• We are pleased that the EBA included the diversification element in the prudential filter.Do you agree that category level AVAs described in Articles 11 to 16 within the core approach should be aggregated as a simple sum? If not, what other approach could be prescribed? State your reasons.
• No not a simple sum but also a diversification benefit of like 25% as there will be overlap between for example operational risk and model risk. Model risk can be viewed as one of the operational risks.Do you agree with the requirement for institutions using the core approach to implement the above ongoing monitoring tool as an indicator of the adequacy of data sources of valuation inputs used to calculate the AVAs described in Articles 8 to 10? If not, what other approach could be prescribed? State your reasons.
• We do not agree. We consider this not the correct tool to ascertain the adequacy of data sources or valuation inputs. Ongoing monitoring will result in a huge increase of the administrative requirements and a significant increase in the amount staff. We consider the added value of back testing for liquid positions not useful and for illiquid positions not practical.• An alternative approach could be to state that banks needs to evaluate yearly the liquidity of their positions and substantiate why they consider certain positions liquid due to independent ‘bona fide’ quotes. In addition it would be better that banks are required to document how they have substantiated the fair value of less liquid positions (i.e. what information and judgment was used) that is reviewed periodically by supervisors and auditors. This will enhance the internal controls within banks significantly.
• Additionally, from a practical point of view ongoing monitoring will not be implementable and will require a significant amount of IT investments and human resources that will go beyond the ‘theoretical’ advances that EBA proposes.
• With respect to Article 20 . In particular paragraph 2 is not entirely clear on what is required. A worked out example would be welcomed. As far as the requirement is clear we don’t deem it realistic. Prudential valuation assessment cannot be a daily exercise. As such comparing transaction prices which will be based on daily inputs to prudential valuation inputs maybe meaningless. The as for prudential adjustment corrected contractual prices would be based on numerous prudential valuation inputs. As such comparing contractual prices would not signify much, besides that a prudent valuation would always be less than the contractual prices. It would also create a considerable operational burden. There should be some thresholds in place and restrictions so that for example plain vanilla transactions (FX forwards) are excluded.