Response to consultation on draft Regulatory Technical Standards on valuation
Go back
Where the valuer does decide to challenge assumptions and significant deviations are found, this should be clearly documented and the basis for the valuer’s decision clearly evidenced based on existing accounting and / or prudential rules. However, rather than prescribing a list of types of information to provide (cash flows/discount rates), we propose that the RTS should require the valuer to clearly document these significant deviations in a detailed way, supported where possible by actual economic and market data (or, in the case of a departure from prudential rules, pre-existing supervisory feedback). This would allow the valuer to attribute changes in asset value to whatever underlying value drivers (e.g. default rates, credit spreads) he/she deems appropriate.
More broadly, the EBA should provide greater guidance in Part II on the extent to which the valuer should include or exclude market factors, e.g. stress factors that originate from the market’s perception that the entity may be subject to resolution. This is particularly important in relation to valuation 1, to avoid a self-fulfilling prophecy from market changes due entirely to expectations around resolution.
Given the need to take into account the circumstances during which the entity is subject to resolution, in particular wider market factors, we would also like to add under paragraph 1 the following additional points of uncertainty:
(f) During times of low transaction volumes, significant disparity may emerge between marking to market and marking to model. The valuer must exercise appropriate judgment under such circumstances.
(g) Owing to the threat of resolution there may be a general aversion to transacting with the entity on normal commercial terms, implying a different value for assets held by the entity relative to the value at which other market participants may transact.
(h) Instruments that were previously marked to market but do not have an active market as at the valuation date. As the instruments were previously marked to market, the valuer may not have the benefit of using a robust model.
(i) Level 2 Instruments for which relevant market data is not available. For instance, instruments subject to counterparty credit risk where the relevant credit spread data is not available.
Instead, the use of the buffer is more likely to be firm-specific in nature - e.g. if the level of uncertainty associated with potential losses stems from concerns about the specific entity’s deficiencies in policies, procedures and systems. Imposing a “one-size-fits-all” buffer would not be appropriate for firms with robust procedures and systems in place. Any buffer applied should also therefore be proportionate to the risk areas identified - e.g. if there is high fraud risk in an isolated business line or entity, the buffer should be applied only to the valuations associated with that business line or entity.
More broadly, we strongly encourage the EBA to do more in this section of the RTS, Part III, to require the valuer to coordinate with the resolution authority (where the two are not the same) and ensure that the work being done under the valuation reflects the agreed upon resolution strategy and the relevant tools to be deployed.
In particular, under Article 11, we object to the suggestion that the valuer should present separate valuations for a “sufficiently diverse range of resolution actions that may be adopted by the resolution authority, including but not necessarily limited to actions contained in resolution plans or proposed resolution schemes.” The valuer should only be expected to carry out valuations based on the resolution strategy and - if necessary - the variant strategies identified in accordance with the EBA’s RTS on resolution planning. It is unrealistic - and likely to be unworkable given the valuation timeframe - to expect all potential resolution actions to be examined.
While it is certainly preferable to primarily base equity value on that indicated by actual market transactions in the entity’s shares, in the absence of these or a market price that can be relied upon to reflect a fair and realistic value of the firm’s assets and liabilities, we believe the valuer should estimate the entity’s equity value by using a range of methods (observable market data, fundamental analysis, net asset value etc.) and choose the most appropriate value indication. Each valuation technique has its own set of shortcomings and so relying on a single technique to estimate value would not be appropriate. We would expect a prospective investor to perform a similar analysis, in circumstances where the sale of business, sale of assets, or transfer power were used.
More broadly, we recommend that Article 1 paragraph 2 and the definitions in Article 2 should specify that the reference date of the valuation should be the date the resolution takes place, rather than the date of the decision to resolve a bank. Indeed, the latter may differ from the former, which is more relevant for the purposes of comparing the actual treatment that shareholders and creditors received.
That said, we agree that such flexibility may need to exist in case it is required in specific circumstances, but only if limited to i) information that was publicly available and would have reasonably been available to the management, the resolution authority and the valuer at the time and ii) only if such information would result in a significant negative change jeopardising the effective application of the resolution tools (e.g. if it clearly resulted in a breach of no creditor worse off, requiring compensation to avoid legal challenge).
For the actual proceeds, assessment of difference in treatment per creditor class should be well documented and where relevant, references to national insolvency law or other relevant law, to any observed prices or insolvency proceedings, should be made. However, full documentation of the valuer’s assumptions and underlying value drivers - as outlined above for the hypothetical insolvency valuation - may only be required if the equity value is not already based on independent, observable, market transactions.
Question 2: Should specific types of information be required on deviations from management assumptions, for example on differences in expected cash flows and/or the discount rates?
Yes, subject to the overarching principle that the valuation should be consistent with the relevant accounting and prudential rules. As the RTS suggests, this should only be the case where “significant” deviations are identified - i.e. subject to a materiality level. In such circumstances, there may be a need for further investigation into what led to such deviations (e.g. was there financial mismanagement?)Where the valuer does decide to challenge assumptions and significant deviations are found, this should be clearly documented and the basis for the valuer’s decision clearly evidenced based on existing accounting and / or prudential rules. However, rather than prescribing a list of types of information to provide (cash flows/discount rates), we propose that the RTS should require the valuer to clearly document these significant deviations in a detailed way, supported where possible by actual economic and market data (or, in the case of a departure from prudential rules, pre-existing supervisory feedback). This would allow the valuer to attribute changes in asset value to whatever underlying value drivers (e.g. default rates, credit spreads) he/she deems appropriate.
More broadly, the EBA should provide greater guidance in Part II on the extent to which the valuer should include or exclude market factors, e.g. stress factors that originate from the market’s perception that the entity may be subject to resolution. This is particularly important in relation to valuation 1, to avoid a self-fulfilling prophecy from market changes due entirely to expectations around resolution.
Question 3: Would you add, amend, or remove any areas which are likely to be subject to significant valuation uncertainty?
Under 1.(b) we recommend removing “and the expected evolution of such value after foreclosure” as it may not be possible to predict the evolution of an asset’s fair value, especially those which are subject to significant uncertainty.Given the need to take into account the circumstances during which the entity is subject to resolution, in particular wider market factors, we would also like to add under paragraph 1 the following additional points of uncertainty:
(f) During times of low transaction volumes, significant disparity may emerge between marking to market and marking to model. The valuer must exercise appropriate judgment under such circumstances.
(g) Owing to the threat of resolution there may be a general aversion to transacting with the entity on normal commercial terms, implying a different value for assets held by the entity relative to the value at which other market participants may transact.
(h) Instruments that were previously marked to market but do not have an active market as at the valuation date. As the instruments were previously marked to market, the valuer may not have the benefit of using a robust model.
(i) Level 2 Instruments for which relevant market data is not available. For instance, instruments subject to counterparty credit risk where the relevant credit spread data is not available.
Question 4: Should the buffer instead always be greater than zero? If yes, how should the buffer be determined?
We agree with the statement “In the absence of facts and circumstances supporting the existence of additional losses, the buffer shall have a value of zero”. The requirement in the BRRD to include a buffer reflects the desire to limit the impact of uncertainties regarding the provisional valuation, but as the rest of the RTS generally requires the most conservative valuation and the use of “value ranges” within limits from the EBA’s prudential valuation RTS under Article 3(3), we consider that these uncertainties about potential additional losses are adequately dealt with. Requiring the buffer to be always greater than zero may therefore amount to double counting. In fact, where it is not warranted this may be unhelpful from a “no creditor worse off” perspective, as it may result in some creditors being bailed-in significantly more than is required.Instead, the use of the buffer is more likely to be firm-specific in nature - e.g. if the level of uncertainty associated with potential losses stems from concerns about the specific entity’s deficiencies in policies, procedures and systems. Imposing a “one-size-fits-all” buffer would not be appropriate for firms with robust procedures and systems in place. Any buffer applied should also therefore be proportionate to the risk areas identified - e.g. if there is high fraud risk in an isolated business line or entity, the buffer should be applied only to the valuations associated with that business line or entity.
Question 5: Do you agree that a valuation of post-conversion equity is necessary to inform decision on the terms of write-down or conversion?
In general, we agree with this statement. However, there may be instances when the terms of the write down or conversion is linked to an entity’s share price that is directly observed in the market. In such cases a valuation of post-conversion equity may not be necessary to determine the conversion terms. That said, as outlined above under Q1 and below under Q6, we are concerned about determining equity value based on market price alone. In theory, the price should be in line with the adjusted net asset value, but in practice, this will not always be the case. Hence, we see value in the RTS describing a non-restrictive range of methodologies for determining equity value.More broadly, we strongly encourage the EBA to do more in this section of the RTS, Part III, to require the valuer to coordinate with the resolution authority (where the two are not the same) and ensure that the work being done under the valuation reflects the agreed upon resolution strategy and the relevant tools to be deployed.
In particular, under Article 11, we object to the suggestion that the valuer should present separate valuations for a “sufficiently diverse range of resolution actions that may be adopted by the resolution authority, including but not necessarily limited to actions contained in resolution plans or proposed resolution schemes.” The valuer should only be expected to carry out valuations based on the resolution strategy and - if necessary - the variant strategies identified in accordance with the EBA’s RTS on resolution planning. It is unrealistic - and likely to be unworkable given the valuation timeframe - to expect all potential resolution actions to be examined.
Question 6: Do you agree with the definition of equity value for this purpose in Article 2 (i)? If not, what changes should be made to the definition? Should the definition be more closely linked to the net asset value determined on the basis of the remainder of valuation 2 adjusted for goodwill/’badwill’, and if so how should that adjustment be estimated?
As outlined above, we believe greater flexibility is needed for the valuer in selecting the appropriate methodology to determine equity value, depending on the circumstances. In principle, the market price for transferred or issued shares should be in line with the adjusted net asset value, if independent, third party investors perform a similar ‘economic value’ analysis required by valuation 2 prior to investing in the subject entity’s shares. However, in practice, we recognise this is not always the case, and so in those circumstances the definition under Article 2(i) would not be appropriate. Depending on the resolution strategy, another methodology may also be more appropriate.While it is certainly preferable to primarily base equity value on that indicated by actual market transactions in the entity’s shares, in the absence of these or a market price that can be relied upon to reflect a fair and realistic value of the firm’s assets and liabilities, we believe the valuer should estimate the entity’s equity value by using a range of methods (observable market data, fundamental analysis, net asset value etc.) and choose the most appropriate value indication. Each valuation technique has its own set of shortcomings and so relying on a single technique to estimate value would not be appropriate. We would expect a prospective investor to perform a similar analysis, in circumstances where the sale of business, sale of assets, or transfer power were used.
Question 7: As an alternative, should the use of information that becomes available after the resolution date be more restricted, and in particular permitted only if it refers to facts and circumstances existing at the resolution date which could reasonably have been known at that date?
In general, information that becomes available after the valuation date should not be used when performing valuation analysis. During times of financial crises, low market volumes and heightened uncertainty may lead to significant swings in market data that makes the valuation date an important input in the analysis. In order to make a ‘like for like’ comparison between the actual proceeds received by the shareholders and creditors and the hypothetical proceeds under an insolvency, the valuer should only use information that was available as at the resolution date. Given that the actual proceeds received are dependent on information that was publicly available at that time, the hypothetical proceeds under insolvency also ideally only use information available as at that date. “Hindsight” - e.g. about the evolution of the market after the resolution date - would not change the treatment at that given point in time.More broadly, we recommend that Article 1 paragraph 2 and the definitions in Article 2 should specify that the reference date of the valuation should be the date the resolution takes place, rather than the date of the decision to resolve a bank. Indeed, the latter may differ from the former, which is more relevant for the purposes of comparing the actual treatment that shareholders and creditors received.
Question 8: Should the use of information available after the resolution date be further limited, for example by requiring that such information is only used if it results in a significant change in the values of the entity’s assets or liabilities?
We disagree in general that information available after the date the resolution takes place should be used. The valuation will already seek to take into account quantitative and qualitative factors that may lead to uncertainties, based on publicly available information at a given point in time. We do not believe that “hindsight” should play a role - especially with regards to expectations regarding stressed market conditions, as nobody at the time could have reasonably anticipated how these would evolve.That said, we agree that such flexibility may need to exist in case it is required in specific circumstances, but only if limited to i) information that was publicly available and would have reasonably been available to the management, the resolution authority and the valuer at the time and ii) only if such information would result in a significant negative change jeopardising the effective application of the resolution tools (e.g. if it clearly resulted in a breach of no creditor worse off, requiring compensation to avoid legal challenge).
Question 9: Should these technical standards provide further detail on the characteristics of appropriate discount rates?
We believe it would be beneficial for the RTS to provide further detail, to ensure that the valuer provides clarity as to how the discount rates have been derived. As discount rates should generally reflect the uncertainty associated with cash flows, we would expect a higher degree of uncertainty for an entity under insolvency. Furthermore, this uncertainty may also be dependent on the relevant national insolvency procedures. Consequently, details on the various factors a valuer should consider in coming up with an appropriate discount rate would certainly be helpful, as it will be an integral part of the valuation exercise for the valuer to provide reasoning for their decisions.Question 10: Are there any changes you would suggest to the methodology for determining actual treatment of shareholders and creditors in resolution? In particular, should the methodology for valuing equity be further specified and, if so, what should be included in that specification (whether additional detail on the current approach, or a different approach, linked for example to net asset values adjusted for goodwill/badwill)?
As stated in answer to Q1 and Q6, we believe the RTS should provide flexibility for the valuer to use a range of methodologies to determine equity value, depending on the circumstances and nature of information available.Question 11: Should the valuer be required to accompany the comparison envisaged in Article 7 of this Regulation with additional relevant disclosures? If yes, what should those be (for example, documentation of any differences between the valuation of actual treatment and the market price that would be observed for those same claims were they traded in an active market)?
We believe the valuer should publish a report setting out all of the assumptions underpinning their methodology for valuation, including the underlying data sources, value drivers and major assumptions used. Given the hypothetical nature of determining proceeds under insolvency, we believe this valuation is more subjective than that of the actual proceeds and so requires such disclosures.For the actual proceeds, assessment of difference in treatment per creditor class should be well documented and where relevant, references to national insolvency law or other relevant law, to any observed prices or insolvency proceedings, should be made. However, full documentation of the valuer’s assumptions and underlying value drivers - as outlined above for the hypothetical insolvency valuation - may only be required if the equity value is not already based on independent, observable, market transactions.