Primary tabs


No, we don’t.
The European Commission has released in December 2021 the Directive 2021/2167 relating to credit servicers and credit purchasers, whose article 16 dictates the implementing technical standards (ITS) for Data Templates, and it invites Member States to ensure that credit institutions adopt these revised Templates for submitting this information to prospective investors during NPL disposals. The new regulatory framework stems from the need to simplify and harmonize the discipline of non-performing loans with the aim of creating a transparent and efficient secondary market.
On the one hand, ABI agrees that these Draft ITS can be of great help in improving the process of non-performing loans disposals in secondary markets, ensuring greater protection of the debtor and the buyer; on the other, ABI believes that additional improvements could be made aiming to better balance the increased costs incurred by the sellers for the new data extraction with the benefits obtained by the buyers.
ABI therefore recommends to focus on significant and crucial data fields rather than increasing the number of data fields that do not add significant value for the investors. Furthermore, ABI suggests keeping the possibility of “no data options”, knowing that investors offer a price based on the information received and a significant role of the process is based on due diligence and legal aspects /contracts of the transactions provided during the process.
Analyzing the templates in detail, we see the following criticalities:
- The structure of the overall framework of templates should be reviewed. Currently guarantees/mortgages and collaterals are in the same sheet while they should be separated into two templates, one for guarantees/mortgages and one for collaterals. In particular, mortgages in Italy are not at asset level but at counterparty level. Also judicial procedures should have a dedicated sheet and not be included in the counterparty sheet.
- The number of mandatory fields is still too high, the loan data tape should include the key data to kick off an evaluation of NPLs. Some of the proposed mandatory fields are irrelevant for NPLs and some are relating to public data that the investor could collect bearing the relevant cost if needed, otherwise not only the bank would bear the cost but it would also be obliged to collect such data for all the loans included in the portfolio, whereas the investor would collect such information only where deemed necessary.
- The existence of a dichotomy between the Directive 2021/2167 (Article 16 paragraph 3) - where in relation to proportionality it is explained that the draft implementing technical standards shall be proportionate to the nature and size of credits and credit portfolios - and the EBA proposal for the draft ITS - where proportionality principle is declined only for type and size of credits and not for credit portfolios.
- The threshold of 25,000 euros is irrelevant if the number of mandatory fields were decreased to the minimum as requested in first point. Furthermore. we believe that the proposed threshold should be considered at portfolio level (for a portfolio largely made up of NPL below €25,000, it would not make much sense to impose the use of the extended NPL transaction templates if only a few NPLs exceed the threshold). Nevertheless, should this pragmatic approach (adapted to market practices and to economic reality) be disregarded, which we would highly regret, then we would propose a significantly higher threshold to avoid an unnecessary and costly burden bringing very little added value.
- The tape should not be requested for single names, since market practice does not consider them. For this reason, from ABI’s point of view, the Data Templates should have application in case of loans’ portfolio transactions.
- There is a material risk of creating a disincentive to disposals for smaller banks that would require to do IT investments as well as for plain vanilla transactions like consumer loan portfolio sales that are statistically evaluated by investors based on few key data points and that now would require additional irrelevant data.
- Current framework does not take into account local legal requirements (e.g. on mortgages).
- Due consideration is not given to the time required for implementation of this new mandatory templates. In order to have time for adapting the banks IT reporting systems and be able to meet the requirements set out in the final Draft Technical Standards, ABI recommends the postponement of the entry into application date of the ITS and the adoption of appropriate phase-in arrangements for credit institutions, especially with regard to UTP portfolios which could require a longer testing phase since the number of such transactions in the market is still limited compared to bad loan portfolio sales and sellers might need time to implement the EBA templates for such asset class. Although ABI and other banking associations had highlighted the need for a phase-in arrangement in their responses to the consultation on the EBA Discussion Paper on the review of the NPL Transaction Data Template (EBA/DP/2021/02), we noted that EBA did not analyse potential related costs and benefits of this option on the Impact Assessment (IA) Section 5.2.
ABI emphasizes that a very clear example of the hurdles related to data standardization is the experience which have so far accompanied the adoption of the templates for securitized performing loans required by the European Central Bank and by the ESMA: such standardisation took several years to be reached on performing assets which, for instance, did not need specific information on judicial proceedings
Given the above considerations, and also reflecting the fact that the adoption of the reporting requirements proposed may require substantial time and effort and that collective benefits of a transitional period outweighs the costs for market participants in terms of making the necessary adaptation of their reporting systems, ABI strongly supports the assumption of such phase-in approach over at least 15-18 months. Such proposed duration is based on past experience, as for example when the ECB established its loan-level disclosure templates for ABS in 2013. At that time, market participants were given 15 months to implement the templates, which was then extended by a further 4 months with the introduction of the ‘comply-or-explain approach’.
ABI urges EBA to carefully consider this proposal that, addressing the industry‘s needs, would give a boost to the acceptance and effectiveness of EBA NPL data transaction templates that aim to the goals of transparency and development of functioning secondary markets for NPLs in the EU, with as little extra effort and burden on banks as possible.
ABI believes the Counterparty Template is still too granular and detailed, and not in line with Italian market practice, which is a benchmark for such kind if transactions. There are several data fields on information that is very residual or that may not be present at all on bank’s IT systems. For this reason, being the extraction process too onerous, the costs would overcome the benefits, thus we suggest removing some of those fields or keep them as optional.
Data on balance sheet data are not relevant, are not market standard and should be collected by buyer if needed bearing the related cost (public data). If the bank is obliged to collect it, it would have to do it for all the counterparties while buyers would focus the collection of such data only where relevant for them.
Given these considerations, ABI suggests removing data fields from the mandatory fields – as for example financial statements of the debtors- that could be retrieved by public data sources considering that those data fields have no impact on reducing the claimed information asymmetry. In addition, from ABI’s point of view simplifications and little changes could be made to reduce repetitions in the data fields’ filling-in process.
Finally, with reference to the counterparty's residence data field, ABI points out that only the residence declared and known to the bank at the time of the establishment of the relationship can be reported, with no obligation to check continuously for changes. In most cases, the residence is a field that is reported by province or city rather than by registered address. Residence information does not have a real-time update because in many cases, especially for bad loans, contact with the client is no longer so frequent. In addition, with regard to nationality, also in order not to discriminate against the customer, the simple indication of "resident" or "non-resident" is suggested.
For a detailed and comprehensive feedback on the individual data fields please refer to the comments provided in the dedicated columns of the data glossary of Excel file (Annex II).
ABI believes the Loan Template is still too granular and detailed, and not in line with Italian market practice. Many data fields are not usually exchanged by market players and could therefore be excluded. Moreover, Templates 3 contains some fields that require sensitive information, that it is not appropriate to disclose, also because these data are not relevant for the purposes of data quality and data transparency or are internal evaluations that are not significant from an investor point of view.
The template should be simplified accordingly by:
- avoiding including fields on joint counterparties which should be in the related sheet;
- reducing the number of mandatory fields, in particular there are fields not relevant for bad loans (e.g., loan interest data);
- avoiding fields that can be calculated based on other data (e.g., days of default).
Please refer to the Excel file (Annex II) for specific comments on each data field.
ABI thinks that further streamlining is needed. Indeed, there are cases of data fields demanding information that, beyond being too onerous to be extracted, is currently not in banks’ practices.
The proposed template includes data on guarantees, mortgages, movable/immovable properties and proceedings which in our view should be separated. The proposed template generates various issues:
- a multiplication of rows since a mortgage can be linked to assets linked to other mortgages with different lien, therefore both mortgages and assets can appear different times;
- proceedings are not always related to assets and are usually linked to debtors. This also generates multiplication of rows.
We believe that guarantees/mortgages, assets and proceedings should have dedicated templates.
In addition, some key data is missing, e.g., cadastral data for immovable properties.
Lastly, many fields are mandatory but they provide excessive detail that doesn’t fit the concept expressed in first answer of key basic data, eg. energy classification of the asset.
For a detailed and comprehensive feedback on the individual data fields please refer to the comments provided in the dedicated columns of the data glossary of Excel file (Annex II).
ABI considers that there are some fields referring to past payments that could not be easily extracted, in particular there is the possibility to extract the data mainly for term loans, in fact short term loans can be also revolving without defined collection amounts. In addition, the field relating to the “total repayment schedule” contains sensitive internal information that the banks cannot disclose since the expected repayments have to be determined by the investor during its due diligence process based on its own estimates. Our member banks do not see any benefit about the proposed split between internal or external collections.
Details on the external servicer recoveries are not needed and not in line with market standard. We believe that the sheet should report the recoveries pre cut-off and post classification as mandatory fields while the indication of whether the recovery was internal or external should be non-mandatory. The IT systems do not collect such data and if a debtor has had various servicers, the details of the collections pertaining to such different servicers would be very difficult to retrieve.
Also, there are no data relating to the type of extrajudicial agreement in place with the debtor (repayment plan, dpo), which instead should be foreseen.
Repayment plan schedule should not be mandatory, but the date of the agreement, the amount agreed, and deadline of the agreement should always be provided. The scheduling only if available.
Info on the external servicer could be in the counterparty or loan sheets.
Collection from asset sale is not always tracked as such in the IT systems.
Please refer to the Excel file (Annex II) for specific comments on each data field.
ABI agrees that the structure of Template 2 represents the relationship across the templates.
Protection identifier should be the mortgage or the guarantee, not also the collateral. Collateral instead should have its own identifier.
In addition, it is still not clear the exact meaning of the term “tenant”.
For a detailed and comprehensive feedback on the individual data fields please refer to the comments provided in the dedicated columns of the data glossary of Excel file (Annex II).
ABI believes that the descriptions of the various fields could be more detailed taking into account the legislative differences that characterize the various EU markets. Consequently, from ABI’s point of view, the data glossary (Annex II) should have country tab link for all the fields that have specific/different description from one country to another by giving details for those fields according to the country of reference. Moreover, it could be useful to distinguish the various non-performing loans subject to disposal by asset class and risk category (UTP vs Bad loans).
ABI welcomes the efforts made in building the data glossary, to the extent possible, on existing common EU definitions set out in the EU regulatory, supervisory, and reporting framework (FINREP), the European Central Bank’s AnaCredit and the ESMA templates used for the NPL securitisation purposes. Anyway, in ABI’s view, the fine-tuning of the NPL data templates should be accompanied by a comprehensive rationalisation of data requests on NPLs.
As also recognized by the Directive (EU) 2021/2167, the EBA should be particularly mindful of any duplication and overlap with the existing requirements on NPLs. To the same end, ABI urges a comprehensive rationalization of data requests on NPLs. Following the principle of “report once” shared by the European Commission and the banking industry and several Recommendations of EBA Cost Compliance Study, it is important to fully rationalise the different NPL information sets in order to enhance data comparability, curb overlaps, streamline and increase efficiency in the reporting processes, facilitate data sharing, advance coordination among authorities and reduce undue reporting burdens.
Given the above considerations, ABI suggests the EBA to define the final templates also in light of future changes of the IREF, which will replace all Statistical Regulations, and with the Integrated Reporting System (post EBA Feasibility Study, Article 430c of CRR2).
Please refer to the Excel file (Annex II) for specific comments on each data field.
We believe that the instructions could be more exhaustive and clearer in order to properly fill in the Templates.
Clarifications are needed also regarding the “appropriate internal governance arrangements that are similar to the arrangements put in place in the credit situations, for example, for supervisory or other regulatory reporting”, considering that moreover, it is stated that “the draft ITS does not introduce any supervisory reporting requirements but they shall be used for the exchange of information by the parties potentially involved in a NPL transaction”. ABI believes that the building and implementation of such processes will be a burdensome activity with a negative time to market impact.
Furthermore, clarifications are needed on the one hand regarding “the relevant management with sufficient degree of seniority that may commit the credit institution to the responsibility for the completeness, consistency and accuracy of the information provided” and on the other hand regarding the “appropriate managerial approval process confirming that the credit institution is responsible for the completeness, consistency and accuracy of the information provided”.
ABI urges that further specifications have to be provided in relation to the “sufficient degree of seniority and appropriate managerial approval” stated in the draft ITS.
Regarding the use of “No Data options” for non-mandatory fields ABI supports the possibility given by the EBA to leave some data fields blank, using one of the four options. As noted before, we consider the no-data option of significant importance since the absence of a specific data should not be an obstacle when a potential buyer is willing to price an NPL portfolio. Moreover, the availability of data can also depend on the specific type of loan under consideration.
If banks have the information under the format specified in the NPL template, they will provide it. In the case that they have the data under a different format, we will provide it separately. As already mentioned, the sole use of ND4 for the mandatory fields would be a severe issue for banks/credit institutions leading to slow down significantly or stop any future NPL sales for a certain period of time (at least the time for IT developments). For this reason, ABI urges EBA to extend the use of all “No Data options” to mandatory fields.
Indeed, the impossibility of leaving some mandatory fields blank, unless not applicable (as provided by the fourth option), entails providing inaccurate information or the need for internal evaluations by the buyer. Indeed, as already mentioned, it is in the banks’ best interest to establish long term relationships with the investors. For this reason, banks are committed to improve information asymmetry by respecting and guaranteeing data quality and data transparency, as to negotiate with the buyers a fair price in line with market standards. This belief is agreed upon with other European banks, as pointed out in EBA’s public hearing held on the 15th of June.
We understand, from the definition of ‘ND2’, that the EBA implies that all data come from banks’ reporting systems. What is the definition of reporting system? Many data actually come from operational systems (=business systems), e.g., all data extracted during the month as opposed to some reporting system that only provides a picture at month-end. We would like to draw EBA’s attention on the fact that all mandatory reporting information is not available at any time during the month, but most often, at month-end, although NPL sales do not all occur at the end of month.
In order to make standardization feasible, the templates structure and the mandatory fields should be in line with market standards. Further it should be considered that some fields might be mandatory for certain classification status and not for other ones (e.g., interest on loan in case of bad loans are not required). As already pointed out, the set of Templates should be reduced and we insist on the “no data option” for the various fields. Furthermore, the legislative differences between Member States have to be indicated in the Templates.
First of all, we suggest removing the dichotomy between the Directive 2021/2167 (Article 16 paragraph 3) and the EBA proposal for the draft ITS.
Secondly, we don’t see great value in the split, also because we are downgrading several mandatory fields requested above threshold to not mandatory.
Please also refer to our response to Q11. In summary, we believe that the proposed threshold should be considered at portfolio level (for a portfolio largely made up of NPL below €25,000, it would not make much sense to impose the use of the extended NPL transaction templates if only a few NPLs exceed the threshold). Nevertheless, should this pragmatic approach (adapted to market practices and to economic reality) be disregarded, which we would highly regret, then we would propose a significantly higher threshold, to avoid an unnecessary and costly burden bringing very little added value.
The draft ITS set out requirements for the internal governance arrangements around the provision of data specified in the templates. To this end, the information provided to prospective buyers in accordance with this draft ITS are to be subject to pre-defined internal governance arrangements. We do not support the proposal in the draft ITS with the three-step approach which introduces a fragmentation among different internal functions in the context of the validation process. We remind that sellers have the contractual obligation to provide accurate data. In addition, data accuracy and quality are essential in building our reputation as a seller. It should also be noted that banks have already well-established and effective internal governance arrangements and staff with years of experience whose expertise should not be broken up and dispersed within institutions. Therefore, there is no need to impose a specific and formal process. The process proposed in the draft ITS is not market standard and would lead to impossibility to finalize deals in the usual time to market. It has also to be considered that a datatape is provided to the investor at a specific cut-off date and such data cannot be changed several times during the process for any minor change which might occur. In case a relevant data is not accurate, the bank and the investors always agree on some specific price sensitive fields to be guaranteed. For further details and comprehensive feedback please refer to our response to Q8.