Initiative on an Integrated EU Covered Bond framework COM(2018)94 and COM(2018)93 Intesa Sanpaolo comments Intesa Sanpaolo supports the set-up of a single and holistic European framework for the Covered Bonds as a building step of the Capital Markets Union. Covered Bonds have gained in the last ten years a leading role throughout the EU as a secured funding channel, showing a remarkable resilience even during the most severe phase of the financial crisis. The following table provides a fresh glance on the use of Covered Bonds by Intesa Sanpaolo to implement its funding strategies and maintain its regulatory liquidity indicators. According to Intesa Sanpaolo, a limited number of provisions of the current regulatory draft could trigger unwarranted discriminatory effects in certain jurisdictions and on SPV-based Covered Bond structures. In particular, our concerns are on: Derivatives: hedging interest rate or currency risks should make the covered bond structure sounder. However, we fear that including the derivatives inside the cover pool may work for certain jurisdictions but may not work at all for others. Therefore, the provisions on derivatives should be improved in order to be applicable to all jurisdictions and to avoid systemic risks (by concentrating derivatives on highly rated institutions only). Liquidity Buffers: the Commission s proposal may ultimately increase concentration of liquidity on highly rated institutions, thus discriminating sound banks established in more peripheral jurisdictions. The proposal should be calibrated in order to take into account actual risk factors and the current state of play of the market. Eligible assets: Intesa Sanpaolo believes that the Directive should aim at protecting the covered bond franchise by allowing high quality and traditional assets as collateral. This objective could also be achieved by explicitly introducing a category of new dual recourse securities backed by non-core high quality assets (SME, infrastructure, the so called European Secured Notes), thus also encouraging the financing of the real economy along with the CMU guidelines. 1
Main concerns of the proposed framework 1. Derivatives (Directive, Art. 11 and Regulation Art. 129) high risk of (unwarranted) fall-outs on SPV-based Structures and discriminatory consequences on non-core EU countries Directive: art 11 - Derivative contracts in the cover pool Regulation: art 129 (c) Art. 11 Directive Member States shall lay down rules for cover pool derivative contracts including at least: (a) the eligibility criteria for the hedging counterparties; Art 129 (c) Regulation "(c) exposures to credit institutions that qualify for the credit quality step 1 or credit quality step 2 as set out in this Chapter Implied Issue in the current regulatory draft: the purpose of the aforementioned provisions, as derivatives contribute to the coverage requirement, is to make the covered bond structure sounder requiring that asset swaps are segregated inside the cover pool and specifying i) counterparty eligibility criteria; ii) limits on the amount of derivative contracts in the pool; iii) necessary documentation on derivative contracts. In addition, the link to the CRR implies that to gain a preferential treatment the covered bonds must be collateralised by exposures to credit institutions that qualify for the credit quality step 1 or credit quality step 2. Intesa Sanpaolo view: In our view Art. 11 makes the whole proposal ambiguous and potentially less sound. In fact, derivatives can currently be included in the cover pool in certain jurisdictions only. In other jurisdictions derivatives are only a hedging instrument on the cover pool, but are not part of the pool itself. Therefore, derivative contracts may but should not be included in the covered pool. Considering derivatives inside the cover pool may, in addition, trigger higher volatility (the mark-to market component of the derivatives may increase or decrease the overall value of the cover pool). 2
Please consider that under Italian CB framework, the payments under the swap are segregated in favour of the CB holders. Therefore, in case of insolvency of the issuing bank or in case of insolvency of the swap counterparty such flows cannot be claimed by any third party (which is external to the CB programme and is not part of the relevant documentation eg. Intercreditor Agreement). Therefore, according to the Italian Framework, even if derivatives were included in the cover pool they would not be safer nor make the CB structure sounder compared to the current situation. The provisions contained in the amendment to the CRR may also lead to unwanted consequences: it would in fact restrict the derivative contracts to a very limited number of eligible counterparties, paving the way for an unwarranted and unnecessary systemic risk and increasing the all-in cost of the programmes. Operational consequences: it would be extremely difficult to find an eligible CQS1/CQS2 swap counterparty for large CB programmes (such as the ISP CB Programmes with assets to be hedged: in excess of Eur 50 bn). This would imply to find a number of different and available counterparties) with i) operational and credit issues, ii) heavy impacts on contracts agreement and iii) a non-negligible effects on funding costs; given the extremely limited number of eligible counterparties a new relevant systemic risk would emerge at EU level where hundreds/thousands of billions of derivatives would be concentrated in the hands of few financial counterparties; in ordinary banking practices we should set-up appropriate guaranteed lines with all the third counterparties, denting the room of manoeuvre of the bank as a whole in respect of the selected counterparties (derivatives lines would reduce the ability of the bank to open with these counterparties other credit lines related to different deals); third eligible counterparties would acquire a deep view on a significant part of the banking book in terms of prepayment rate/delinquencies/defaults/retention strategies; an increasing number of structures could be deprived of hedging structures to limit the above described effects of externalising the role. Would be the unhedged structure safer for the investors compared to the hedged ones? A very efficient alternative solution would be to provide that derivative counterparties can qualify also for the credit quality "step 3". This would allow all investment grade counterparties to be eligible, taking into account that the lower the rating of the swap counterparty, the larger the collateral amount required under Rating Agencies criteria (with the same methodology of the Rating Agencies setting the minimum acceptable rating threshold to be eligible). Intesa Sanpaolo proposal of a wording amendment: Art 11: 3
<<Article 11 Derivative contracts in the cover pool Derivative contracts entered into in the context of a covered bond programme shall ensure investor protection and have to meet all the requirements below. Member States by allowing may allow derivative contracts to be included in the cover pool only where at least the following requirements are met: The derivative contracts are entered into included in the cover pool exclusively for risk hedging purposes; a. ( ), b. Only if included in the cover pool, the derivative contracts are segregated in accordance with art. 12 c. The derivative contracts cannot be terminated upon the insolvency or resolution of the credit institution issuing covered bonds; d. ( ) >> art 129(c) Regulation: (c) exposures to credit institutions that qualify for the credit quality step 1, or credit quality step 2 or credit quality step 3 as set out in this Chapter. 2. Liquidity buffer (Directive, Art. 16) high risk of (unwarranted) discriminatory consequences for non-core EU countries Directive Art 16: requirement for a cover pool liquidity buffer (3) Member States shall ensure that the cover pool liquidity buffer referred to in paragraph 1 consists of the following types of assets: (a) assets qualifying as level 1, level 2A and level 2B assets pursuant to Articles 10, 11 and 12 of Delegated Regulation (EU) 2015/61, valuated in accordance with Article 9 of that Delegated Regulation and segregated in accordance with Article 13 of this Directive; (b) exposures to credit institutions that qualify for the credit quality step 1, in accordance with Article 129(1)(c) of Regulation (EU) No 575/2013. Implied Issue in the current regulatory draft: Liquidity buffers cover the net liquidity outflows of the covered bond programme over the 31-180 days range. Liquid assets are Level 1, 2A assets and Level 2B as well as exposures to credit institutions qualifying step 1. 4
Intesa Sanpaolo view: From a mere technical perspective: point a): as all European banks are subject to liquidity requirements according to the paragraph 4 of art. 16, the already existing Level 1, Level 2A and Level 2B assets should be considered for the current provision; point b): It is not clear why institutions qualifying for step 1 only should be considered eligible. This seems inconsistent with the rationale underlying the previous a) provision. Operational consequences: given the limited number of eligible counterparties a new relevant systemic risk would emerge; high-cost related to the externalisation of the role that would impact the non-eligible banks only (mainly non-core countries banks). In a negative-yields environment the externalisation of the account bank where the liquidity buffer is expected to be held will result expensive for the originator/issuer, thus introducing regulatory benefits for high-rated financial institutions. Intesa Sanpaolo proposal of a wording amendment: Art. 16 Directive (3) Member States shall ensure that the cover pool liquidity buffer referred to in paragraph 1 consists of the following types of assets: (a) (.) (b) exposures in form of cash deposit to credit institutions of any credit quality steps to the extent that, in the relevant jurisdiction, they are segregated. 3. Eligible assets (Directive, Art. 6) Member States shall ensure investor protection by requiring that covered bonds are at all times collateralised by high quality assets referred to in points (a) to (g) of Article 129(1) of Regulation (EU) No 575/2013 or by other high quality assets that meet at least the following requirements: (a) either the market value or mortgage lending value of the assets can be determined; (b) a mortgage, charge, lien or other guarantee on the asset is enforceable; (c) all legal requirements for establishing the mortgage, charge, lien or guarantee on the asset have been fulfilled; (d) the mortgage, charge, lien or guarantee securing the asset enable the credit institution issuing covered bonds to realise the value of the asset without undue delay. 5
Implied Issue in the current regulatory draft: Covered bonds could be collateralised by high quality assets different from those referred to in Art 129(1) CRR points (a) to (g). It s not clear if a new regulation for European Secured Notes (ESNs) will be proposed. Intesa Sanpaolo view: due to the broad scope of Article 6 and the room for interpretation in the wording of recital 15, as well as the legislative provision indicating the high qualitative features, we would propose to reconsider the introduction of the European Secured Notes concept. The introduction of a separate instrument, the ESN, would be consistent with the current provisions of Art 6 and would prevent a watering down of the qualitative scope of the covered bond label. At the same time, it would also be fully aligned with the proposal of the Own Initiative Report of the European Parliament. In such case, it is necessary to recognize in the Regulation a preferential prudential treatment for ESNs, different from the treatment recognised to covered bonds. Operational consequences: the traditional covered bond franchise would be appropriately differentiated and protected by law, as SME/infrastructure/non-core assets would fall in the scope of the ESN instrument; the revamp of the ESN initiative would be strategic for the EU financial system; enlarging the eligible assets for a double recourse instruments (with clear distinction between CB and ESN) could reduce the cost of funding of commercial banks, paving the way for further cuts in spreads applied to commercial borrowers; optimization of the commercial assets otherwise useful for ABS SME. Intesa Sanpaolo proposal of a wording for the introduction of a new Art. 6 - bis: Art. 6 - bis 1. Member States may allow credit institutions issuing dual recourse debt instruments covered by different assets than those required for covered bonds, labelled European Secured Notes (ESNs). 2. EBA lays down the minimum requirements that ESNs covered assets have to meet. 3. The Regulation (EU) 575/2013 allows for a preferential treatment of ESNs. 6
4. Additional topics Composition of the pool (Directive, Art 10) Member States shall ensure investor protection by providing for a sufficient level of homogeneity of the assets in the cover pool so that they shall be of a similar nature in terms of structural features, lifetime of assets or risk profile Implied Issue in the current regulatory draft: given the nature of Programme, the cover pool of each structure is enriched periodically through the transfer of new eligible assets by the Originator. Over the time the financial features of the mortgage loans may therefore change: we expect that such evolution of the cover pool over time will not be considered by the new regulation as non homogenous. Intesa Sanpaolo view: the current provision states that homogeneity should be respected in terms of structural features lifetime of assets risk profile according to our experience all the above elements should be carefully analysed in order to avoid introducing eligibility criteria hard to comply with. For instance, in Italy, the existing law requires that the eligibility criteria that define the transferred pool should be reckonable by the affected borrowers. Some criteria related to the risk profile could not be verified by the borrowers (that do not know their valuation resulting from the internal rating model). In addition, we have some substantial doubts about the meaning of risk profile : it is clear that mortgage loans with indexed LTV in the 60-70% range are riskier than those in the 20-30% bucket but both of these pools are perfectly consistent to the existing (and future) regulatory requirements. Operational consequences: net reduction in transferrable pools; increasing opacity in eligibility criteria; less efficient diversification of the cover pool; increase in funding costs. Intesa Sanpaolo proposal of a wording amendment: Art. 10 Member States shall ensure investor protection by providing for a sufficient level of homogeneity of the assets in the cover pool so that they shall be of a similar nature in terms of structural features, lifetime of assets or risk profile. 7
Level of mandatory overcollateralization (Regulation) We understand the cautious approach of providing a level of 5% with the exception of certain assets which would require a lower amount (2%). Given the set-up of the current provision we understand that the rationale of this approach is a sort of distrust in respect of the assets reported in the point f) and g) of the current Art. 129 which are deemed to be riskier. In general terms, we believe that markets participants (originators, investors and rating agencies) should be able to define in an independent way the inner risk profile of each CB Programme, as long as such risks are duly communicated to the market in a transparent way. Note that the structural feature of the assets are not, per se, the key driver in defining the risk profile of the structure and -consequently- the equilibrium OC. In fact, the two main drivers are: the temporal mismatch between the time when the assets are repaid by the borrowers (becoming cash) and liabilities maturity. Covered bond Programmes show generally a relevant mismatch between the expected life of the assets vs the weighted average life of the covered bonds: independently from the nature of the underlying assets the bigger is this mismatch, the higher is the equilibrium OC; the uplift of the covered bond rating in respect of the Issuer rating. The higher the targeted rating level on the covered bonds versus the Issuer s senior unsecured notes, the higher the resulting equilibrium OC. Finally, please note that the process implemented by Rating Agencies in defining the equilibrium OC takes into consideration the nature of the assets with a bottom-up analysis. The first stage of their process (static analysis) grounds on a loan-by-loan check of the cover pool while the following ones (dynamic analysis) factorise idiosyncratic stresses on the expected cash flows. Hence, riskier assets would in any case result in higher equilibrium OC. The consequence is that required OC levels do not simply derive from the specific asset class. Therefore, we would recommend that such lower percentage (2%) be applicable for all eligible asset classes permitting to the market participants their own valuations based on structural features of each Programme. Here below you ll find the current status of Intesa Sanpaolo s (1) actual OC levels: as you may see factors influencing OC are the rating levels and the WAL mismatch between assets and liabilities. (1) : Intesa Sanpaolo rating (as of 31.03.2018): Baa1(Moody s) and BBB High (DBRS) 8
Contacts Intesa Sanpaolo International and Regulatory Affairs Department Square de Meeûs, 35-1000 Brussels - Belgium Tel +32 (0) 2 640 00 80 - Fax +32 (0) 2 640 26 74 www.group.intesasanpaolo.com 9