The ECB considers the internal capital and liquidity adequacy assessment processes (ICAAPs and ILAAPs) to be core internal risk management processes for institutions in managing capital and liquidity adequacy. Accordingly, the ICAAP and ILAAP are important input factors into the European Banking Supervision’s Supervisory Review and Evaluation Process (SREP). They feed into all SREP assessments and into the processes for determining Pillar 2 capital and liquidity requirements. The ECB intends to further increase the important role the ICAAP and ILAAP play in the SREP assessment. Among other things, both the qualitative and quantitative aspects of an institution’s ICAAP – the latter being the risks the institution has identified and quantified – will play an enhanced role in, for example, the determination of additional own funds requirements on a risk-by-risk basis.
Given the key role of the ICAAP and ILAAP in both increasing institutions’ resilience and providing supervisors with valuable information on institutions’ capital and liquidity situation, the ECB’s experience shows that ICAAPs and ILAAPs need to be improved across institutions. To this end, the ECB initiated a multi-year plan to develop an enhanced set of supervisory ICAAP and ILAAP expectations for significant institutions, in close dialogue with the industry. The ECB Guides to the ICAAP and ILAAP (Guides) set out the ECB’s understanding of the ICAAP and ILAAP requirements for significant institutions as stipulated in the CRD IV. (See Articles 73 and 86 of Directive 2013/36/EU of the European Parliament and of the Council of 26 June 2013 on access to the activity of credit institutions and the prudential supervision of credit institutions and investment firms, amending Directive 2002/87/EC and repealing Directives 2006/48/EC and 2006/49/EC (OJ L 176, 27.6.2013, p. 338).) Significant institutions are encouraged to put into practice the expectations outlined in the Guides. However, it is important to emphasise that these Guides are not legally binding and therefore do not substitute or supersede any applicable (national) law.
With a view to fostering a common understanding of the Guides, the following provides the ECB’s answers to questions it frequently received during its internal and external discussions on the ICAAP and ILAAP.
The ECB’s experience shows that ICAAPs and ILAAPs need to be improved across institutions. Considerable progress has been noted only in some significant institutions’ ICAAPs and ILAAPs.
The ICAAP and ILAAP are of fundamental importance for institutions in managing capital and liquidity adequacy. To promote the improvement of significant institutions’ ICAAPs and ILAAPs, the ECB follows a multi-year plan. Based on an intensive dialogue with institutions and other industry participants on the draft guidance published in 2017 and taking into account further input, the ECB has refined and enriched its supervisory guidance on the ICAAP and ILAAP. It is important to emphasise that significant institutions are encouraged to start putting into practice the ECB’s supervisory expectations outlined in the Guides without delay, even if they will only be in use by ECB supervisors from 1 January 2019. If significant institutions decide to follow the Guides, they are invited do so in close cooperation with their Joint Supervisory Team (JST), informing them proactively of their situations and plans.
It should be noted that the ECB has not changed the overall philosophy of the Guides since the publication of the first ICAAP and ILAAP expectations back in January 2016. It has only clarified those expectations in three steps, namely in February 2017, March 2018 and with the publication of the final versions in November 2018.
The reference date for significant institutions when submitting ICAAP and ILAAP information (including the ICAAP template for example) to their JSTs in 2019 is 31 December 2018. The ECB will use the Guides to assess significant institutions’ ICAAPs and ILAAPs from 1 January 2019 only.
It is expected that the data and information on ICAAPs/ILAAPs submitted to the ECB by significant institutions in 2019 will take the newly published Guides into consideration. It should be noted that the ECB has not changed the overall philosophy of the Guides since the publication of the first ICAAP and ILAAP expectations back in January 2016. It has only clarified those expectations in three steps, namely in February 2017, March 2018 and with the publication of the final versions in November 2018. Furthermore, the Guides are fully in line with the ICAAP and ILAAP expectations published in January 2016. It is important to note that the ICAAP information provided via the ICAAP template should be from the economic perspective.
Owing to the significant differences in the treatment and role of the ICAAP and ILAAP across Member States, significant institutions’ practices appeared to be very heterogeneous. Differences in the ICAAP, for example, were evident, among other things, in the overall role of the ICAAP in institutions’ management and decision-making, in the roles of the economic versus the normative (or regulatory) perspectives and in the general ICAAP approach, i.e. the so-called “going” versus the “gone concern” concepts. In some Member States, the ICAAP and ILAAP are seen as the core of institutions’ risk management processes, while in others, they are associated with the process of preparing a report on capital and liquidity adequacy for the supervisor or used as synonyms for these reports. The ECB has noted considerable progress only in some significant institutions’ ICAAPs and ILAAPs.
All significant institutions are encouraged to improve their ICAAPs and ILAAPs. We expect that the publication of our significantly enhanced Guides will assist them in that regard.
First, the ICAAP and ILAAP are important input factors in European Banking Supervision’s Supervisory Review and Evaluation Process (SREP). They feed into the assessments of all SREP elements and into the Pillar 2 capital and liquidity determination processes. In the future, it is envisaged that the role of the ICAAP and ILAAP in the SREP will be expanded further.
ICAAPs and ILAAPs are key factors for institutions’ resilience. Therefore, the ECB annually assesses these processes as part of the SREP. Should the ECB identify any weaknesses, it may take supervisory measures to address the issues in the respective institution. Such measures may also include capital or liquidity add-ons to address the increased uncertainty in institutions’ risk management and to incentivise significant institutions to address those weaknesses.
The ECB’s view is focused on keeping the banking sector stable by ensuring that institutions maintain the continuity of their operations. Therefore, the ICAAP Guide describes an ICAAP approach aimed at contributing to the continuity of institutions by ensuring their adequate capitalisation. Since January 2016 the ECB has constantly encouraged significant institutions to change from “gone concern” ICAAP approaches to approaches aimed at their continuity.
The ICAAP and ILAAP are both based on two equally important but complementary pillars: the economic and the normative perspective. Both perspectives shed light on risks an institution is exposed to and on its capital and liquidity adequacy from very different angles. To capture the risks that an institution is exposed to, it should consider managing its capital and liquidity adequacy from both the normative and the economic perspective.
Institutions have to fulfil Pillar 1 and Pillar 2 capital and liquidity requirements at all times (as assessed in the normative perspective). Following this perspective only, however, is insufficient for ensuring the survival of institutions at all times, as shown by the recent financial crisis. For instance, some institutions looked healthy from a regulatory capital perspective, yet struggled to ensure sufficient liquidity and funding levels because their counterparties did not have sufficient trust in them and no longer accepted them as reliable counterparts. Those counterparties knew that the economic substance of such institutions had deteriorated, but this was not (yet) reflected in their balance sheets and, for instance, in their regulatory capital ratios. The literature used the term “zombie banks” to describe such institutions.
Therefore, one perspective cannot replace the other. Rather, both perspectives should complement and, in particular, mutually inform each other.
In the past some significant institutions used ICAAP approaches which they referred to as “going concern” approaches. The fundamental premise of such approaches was to see at that point in time (t0) whether they would still meet the regulatory and supervisory own funds demands if the risks they quantified in their ICAAPs for the upcoming 12 months were to materialise. Essentially, for that assessment, they deducted from their current (t0) own funds in the balance sheet the portion which was necessary to fulfil supervisory/regulatory capital demands. They then compared the remainder – sometimes referred to as “free own funds”– with the risk exposure they had quantified. The risk amount contained all risks that could potentially affect the regulatory own funds and Pillar 1 ratios over the next 12 months. Quite often, risk was quantified using models that produced, for example, VaR with a 99% confidence level (in some cases, lower confidence levels were chosen in order to consider the continuity aspect of their business as contrasted with the “gone concern” approach).
In effect, their current own funds were compared with their Pillar 1 plus Pillar 2 demands, plus all risks quantified in the ICAAP (credit risk, market risk, operational risk, IRRBB, etc.) for t0.
The normative perspective, as defined in the ICAAP Guide, is different to such approaches as it does not foresee a separate ICAAP risk quantification for t0. Instead, significant institutions are expected at t0 to determine Pillar 1 ratios and compare them with their external capital requirements (Pillar 1, Pillar 2 capital requirement, buffers) and Pillar 2 capital guidance. This action is conducted again after one year by projecting the Pillar 1 ratios at t1, as well as for the subsequent years, t2, and t3 at least. These projections are naturally expected to take into account all effects that impact the future Pillar 1 ratios under the respective scenarios. This includes changes in risk-weighted assets, alongside P&L and other own funds effects arising from defaulting credits, market price movements, interest rate changes, etc.
The normative perspective is comparable with what many institutions did in their planning of regulatory capital. However, significant institutions should be aware that the ECB’s expectations with regard to capital planning go clearly beyond what many institutions did in the past, for example, in relation to the determination of adverse scenarios and the severity of assumptions regarding future developments assessed in adverse scenarios.
No, that is not the case. However, the ECB does expect significant institutions to be prudent and conservative. This means that they should not be less conservative overall when setting the parameters and other assumptions underlying their risk quantification methodologies in the economic perspective.
This should not be confused with the application of floors by supervisors. As foreseen in the EBA SREP Guidelines, the supervisor will apply a Pillar 1 floor for specific risks.
In general, the Guides are not legally binding and therefore they do not substitute or supersede any applicable law implementing Articles 73 and 86 of the CRD IV. In the event that, in certain instances, the Guides were not in line with applicable law, the ECB would apply the applicable law in its assessment of significant institutions’ ICAAPs and ILAAPs. However, the ECB has developed the Guides in very close cooperation with the national competent authorities. Therefore, it does not expect such conflicts to arise between the Guides and national law.
No. The ECB clarifies in the Guides that this would be a misperception, since the management buffer concept does not actually set new minimum capital/liquidity requirements above the existing legal minima. Although it is generally expected that management buffers will be larger than zero, an institution may also be able to argue that, depending on the scenario assessed, a management buffer of zero would still allow it to sustainably follow its business model. The management buffer concept merely describes the fact that institutions will usually operate above supervisory minimum requirements on their own initiative, simply because they would not otherwise be able to find the counterparties, clients, employees and investors they need to follow their business model. However, the Guides do invite institutions to explicitly assess the level of capital/liquidity they individually need for each of the scenarios assessed. They are expected to determine concrete management buffers, justify them and document the justification. This follows the general spirit of the ICAAP, in that institutions should be fully aware of the risks they are facing and should manage them actively.
The Guides do not describe concrete adverse scenarios under the normative perspective, because these are expected to be commensurate with institutions’ business activities, operating environments, risk profiles and, consequently, vulnerabilities, all of which are very different from institution to institution. The Guides clarify under Principle 7 that “The range of adverse scenarios is expected to adequately cover severe economic downturns and financial shocks, relevant institution-specific vulnerabilities, exposures to major counterparties, and plausible combinations of these”.
Regarding the level of severity expected to underlie adverse scenarios, the Guides clarify in relation to Principle 7 that the ECB understands “adverse” to mean severe stress: “The level of severity is expected to correspond to developments that are plausible, but as severe from the institution’s perspective as any developments that might be observed during a crisis situation in the markets, and cover factors or areas that are most relevant for the institution’s capital/liquidity adequacy”.
The definition of internal capital is expected to be consistent with the economic capital adequacy concept and internal risk quantifications of the institution. The economic capital adequacy concept is an internal concept aimed at ensuring – under the economic perspective – that the financial resources (internal capital) of the institution will enable it to cover its risks and maintain the continuity of its operations on an ongoing basis.
It should be noted that the amounts of internal and regulatory capital can differ significantly, owing to the different concepts underlying the definitions. This is due to the fact that internal capital is expected to reflect the economic value of the institution, while regulatory capital is based primarily on regulatory definitions, but generally may also contain accounting assumptions. Depending on the individual situation of each institution and the applicable accounting standards, economic values diverging from book values can cause significant differences.
The same is true for the quantification of risks where the normative view assesses the impacts of all risks on regulatory ratios, thus following accounting rules and regulatory definitions. On the contrary, the economic perspective assesses how the universe of risks to which the institution is materially exposed may impact its economic value. Credit spread risks for positions that are not booked at fair value are a typical example of risks that are expected to be treated differently under the two perspectives.
The definition of internal liquidity buffers is expected to be consistent with the economic liquidity adequacy concept and internal risk quantifications of the institution. The economic liquidity adequacy concept is an internal concept aimed at ensuring, under the economic perspective, that the financial resources (internal liquidity) of the institution can cover the risks and expected outflows, and maintain the continuity of its operations on an ongoing basis. Here, again, different underlying assumptions and concepts may lead to major differences in the amount of available liquidity and in stable funding sources.
For Tier 2 capital instruments/subordinated debt instruments, there is usually no provision in the term sheet stipulating that they would absorb losses in scenarios other than liquidation. Subordinated loan positions will be repaid to debt holders in the state of continuation, under the issuance conditions/terms.
Consequently, following the assumption of continuation underlying the ECB’s ICAAP expectations, Tier 2 capital instruments, including in particular subordinated debt, can be generally discounted as risk-absorbing in a continuation state. Accordingly, such instruments generally are not expected to form part of internal capital. However, the institution has the option to justify why this logic is not relevant in a particular case.
In adverse scenarios under the normative perspective, the institution is expected to assume exceptional, but plausible developments with an adequate degree of severity in terms of their impact on its regulatory capital ratios, in particular the CET1 ratio. The level of severity is expected to correspond to developments that are plausible, but as severe from the institution’s perspective as any developments that could be observed during a crisis situation in the markets, factors or areas that are most relevant for the institution’s capital adequacy.
Of course, institutions are not expected to “plan” for such scenarios. However, they are expected to prepare for a plausible case in which they could materialise. “Planning” in that sense is not to be understood as aiming at entering into adverse circumstances. Instead, “planning” is meant to be about “being prepared” and “being able to prevent” future stress events in order to avoid an undercapitalisation under those adverse circumstances.
The level of conservatism underlying the risk quantifications in the economic perspective is already expected to be very high and therefore should also capture very rare events. Depending on the risk quantification methodology used, this is reflected, for example, in high confidence levels. This gives rise to the question as to whether additional stress testing is expected in the economic perspective.
The answer to this question is two-fold: on the one hand, the risk quantifications in the economic perspective are not expected to be subject to stress testing via adverse (multi-year stress) scenarios, as in the normative perspective. However, as the ICAAP should capture future developments, institutions are expected to assess the sensitivities of their risk quantifications to potential future economic developments that are not captured by the data used for quantifying risk.
As an example, house prices in the United States rose continuously over a very long period of time before the onset of the crisis in 2008. The purely backward-looking risk quantification methodologies broadly used by the industry suggested that this long-lasting trend would never change, implying that continuously increasing house prices meant that mortgage loans entailed no credit risk. When the trend changed, history showed that this conclusion was wrong.
A similar issue can be observed when, for instance, institutions rely on value-at-risk (VaR) models to quantify their market risk. After a prolonged period of stable/rising stock market prices, using, for instance, the historic simulation concept for quantifying VaR would result in very low risk figures. In reality, however, trends can always change such that the past captured by the models can materially underestimate the true risk. In both cases, even an extremely high confidence level would result in a fundamental underestimation of risks.
Therefore, a comprehensive, sound and conservative forward-looking stress testing programme in the sense of an assessment of varying parameters is expected for the economic perspective too, as this is key to ensuring that risks are not underestimated and that institutions can maintain adequate levels of capital, also from an economic perspective. Following the “mutual information” concept between the two ICAAP perspectives, institutions are expected to also use the normative perspective adverse scenarios when assessing their capital adequacy under the economic perspective.
It should be noted that sometimes the term “stress testing” is used as a synonym for the EBA-style multi-year stress tests. This is not what is meant in this context. The ICAAP Guide clearly says that the ECB does not expect institutions to produce multi-year projections of its economic capital adequacy situation by, for instance, projecting a one-year one-factor model for another three years.
The risk quantification methodologies and assumptions used under the economic and normative perspectives are expected to be robust, sufficiently stable, risk-sensitive and conservative in order to quantify losses that may arise, even in rare circumstances.
In the view of the ECB, in a sound ICAAP the overall level of conservatism under the economic perspective is, generally, at least on a par with the level underlying the risk quantification methodologies of the Pillar 1 internal models. Rather than one by one, the overall level of conservatism is determined by the combination of the underlying assumptions and parameters. “Combination” in that sense means that approaches can be in line with this guidance, despite certain individual parameters, for example, the confidence level of economic capital models being less conservative than in Pillar 1. However, in such cases, institutions are expected to demonstrate how these less conservative assumptions are offset, in order to be, in combination, at least on a par with the level of conservatism underlying Pillar 1.
With regard to the level of conservatism, the ICAAP Guide clarifies that, although it is generally expected that the ICAAP be used in decision-making, it is not expected that each and every business decision be taken on the basis of the highest level of conservatism.
Different levels may be applied for different purposes. However, institutions are expected to be able to also bear the risks quantified based on a very high level of conservatism. As an example, an institution may base the pricing of derivatives and, thus, the decision to offer or buy certain products in the market, on the assumption that this market will not enter into a crisis situation. However, it should be in a position in terms of both available capital and risk management processes to survive any materialisation of that risk. In the specific example, this would mean that it should be able to absorb any losses incurred from that derivative transaction even if the market entered a crisis situation.
The Guides state the expectation that internal reviews are carried out comprehensively by the three lines of defence, including business lines and the independent internal control functions (risk management, compliance and internal audit), in accordance with their respective roles and responsibilities. In order to ensure a sound system of checks and balances, the area (second line of defence) that is responsible for developing and validating risk quantification methodologies has to be an area that is independent from risk-taking units (first line of defence). In addition, it is important that all activities within the institution (including those of the second line of defence) are subject to a regular review by another, fully independent internal audit function (third line of defence), which reports directly to the management body.
Depending on the size and complexity of the institution, various organisational solutions may be adopted to ensure independence between the development and validation of risk quantification methodologies. However, it is expected that the concepts underlying the various lines of defence are respected, i.e. the independent validation is expected to be conducted by a unit other than the internal audit function as such.
Accordingly, the validation activities, which are expected to be conducted by the risk management function (i.e. the second line of defence) are expected to be subject to regular reviews by the internal audit function (i.e. the third line of defence). It is also expected that internal audit includes the adequacy of risk quantification methodologies in its audit plan.
No. The LCR projections for adverse scenarios (i.e. under stressed conditions) follow exactly the provisions regarding weights, run-off rates, etc. in Commission Delegated Regulation (EU) 2015/61, i.e. the way the LCR is calculated is always the same. However, the ECB expects institutions to determine the outstanding balances of assets, liabilities and off-balance sheet commitments that enter the calculation of the LCR during a period of stressed conditions and then multiply those balances by the weights or run-off rates provided for in the Regulation.
The objective of this expectation is that institutions are aware of the impact certain severe, but plausible future developments may have on LCR ratios (which they would also have to calculate in the future under such conditions). Institutions are expected to analyse the outcomes of these projections and decide whether they need to take action to prepare for or prevent the projected situation.
As an example, the outcome of such an assessment could be that the LCR drops to 60%. While the CRR foresees that the LCR may, at times, fall below 100%, the institution would still be expected to answer the question as to whether it would still be able to sustainably follow its business model with an LCR of 60% under the circumstances it has assumed in the respective adverse scenario.
The ICAAP Guide states that “The capital plan is expected to comprise baseline and adverse scenarios and to cover a forward-looking horizon of at least three years”. This means that institutions are expected to implement a capital planning process – often within the scope of its regular multi-year planning process – that captures the full time horizon of at least three years when it is being approved. Institutions are expected to ensure that the capital plan is adjusted throughout the year if it has become outdated by real developments. In addition, institutions are expected to go through the full capital planning process, involving all relevant functions, such as the economics department, risk, finance and business areas, at least once a year.