Response to consultation on Implementing Technical Standards on NPL transaction data templates
Go back
First, the number of mandatory data fields included in the proposed templates is significantly higher than what has historically been necessary to sign and close voluntary NPL deals, including secured and unsecured, residential and SMEs portfolios. This information is too detailed and is not deemed critical for purposes of loan valuations.
Despite a significant decrease of the data fields compared to the last template revision from May 2021, the remaining number is still high, and it contains notedly more information than the current market standard. In other words, it is still seen as too elaborate and granular. Some of the fields even surpass the information that is relevant for portfolio valuations – the general impression is that de-spite the revision and reduction of the previously named “critical fields”, data specifications are still very much statistically driven (evident by the compliance of many data points with values already known from AnaCredit /collected there, that are of a more statistical nature but do not affect the valuation or processing of exposures).
Also, as one of the participants in previous consultation sessions pointed out, there is a highly noticed influence of third parties with interests other than those of the sellers and buyers, as direct NPL transaction participants, which altered the nature of the templates more towards supervisory and sta-tistical purposes, as well as purposes of transaction platforms that have identified the data and its analysis and provision as a source of revenue, and advisors who want to sell their services around data validation.
Moreover, what has been noticed is that a lot of previously labeled “non-critical data” are shifted to category of “Mandatory” data with the latest draft, either without any definition change or with slight changes in field definition (increase from 70+ to over 130 mandatory data).
As a direct consequence of the increase in mandatory data fields, there would not only be a relevant increase in the credit institutions’ IT costs to include some of those new mandatory data fields in the banking systems, but also an increase in the time-to-market of the NPL portfolios, as some of the proposed mandatory fields may not be included in the systems within a reasonable time frame and the sellers would have to do it manually. This would probably mean an entry barrier to new potential entrants that have lower IT budgets.
Secondly, we think that it makes no sense to have common templates for single names or reduced portfolios of single names and massive portfolios of NPLs. Exposures to one single debtor or to a reduced number of corporates or SMEs have historically needed a different set of information, as their potential purchasers perform a deeper financial and legal analysis of the exposures rather than a statistical analysis, which is more adequate for massive portfolios.
Thirdly, the proposed templates would generate data inconsistency, as these would mainly contain management data, which would be incomparable between European banks and it would lead not only to misunderstandings among potential investors, but also to proposed unjustified price adjustments that may result in many transactions not being closed, due to comparing data between heterogeneous EU jurisdictions and practices.
Fourthly, the expected impact on the cost side for the sellers of NPL should be taken into consideration. Namely, there is generally a high correlation between high data quantity, potential for error and greater validation effort. This has been recognized by the EBA, and in the part referring to Internal governance, this added pre-sale validation burden is to be endured by the banks - the ITS draft requires credit institutions to establish an internal process ensuring that the information being provided to the prospective buyers has been validated by staff independent from the staff involved in the sales process and is subject to an appropriate managerial approval. This leads to a due diligence that should be carried out by the bank before it is even known whether a sale will take place (when demand and prices are still unknown) and building additional costs for the banks for data gathering and validation, causing the total transaction cost to increase. In other words, overly extensive data requirements create more complexity, effort, and liability risks – this in particular can create signifi-cant obstacles for smaller banks and might prevent them from entering the market, as they do not have sufficient NPLs in total numbers to amortize the additional cost for data validation.
Furthermore, the data sheet should have a disclaimer stating that the credit institution should not assume any liability for the information to be complete or sufficient for an in-vestment decision of the prospective buyers. The market for distressed assets on the buyer side is dominated by investment banks having specialized units trading and investing in distressed asset and by specialized distressed funds. The professional and analytical level on the buying side is very high. Investing in distressed assets is a high-risk business. We doubt that it would make sense to widen the investor basis in providing a data sheet as a basis for the necessary analysis. For an investment deci-sion the data will regularly not be sufficient. The analysis must be made by the buyer and its risk must be borne by the buyer as laid down, for example, in the LMA standards for distressed asset sales. If that risk partially moves to the seller side the market will not get bigger but significantly smaller. If data like the valuation of securities, or balance sheet information are included, the analysis partially moves to the seller side. Information that includes a valuation or analysis, in our opinion, should therefore not be part of the NPL Transaction Data Template. If certain information is provided in a data room, it should be possible to exclude this information from the data sheet. If for example the complete legal documentation is provided for the purpose of a legal due diligence by potential inves-tors, a data sheet summary of the legal structure should be obsolete.
Additionally, the aforementioned reasons imply that the proposal is misaligned with article 16(4) of the Directive, which states that, when preparing this draft ITS, both existing market practices in data sharing between buyers and sellers, as well as minimizing processing costs for credit institutions and credit purchasers, should be taken into consideration.
In conclusion, we would like to request that the EBA further streamlines the templates, aim-ing at simpler, more balanced and effective design in order to achieve a broader application and in-crease transparency in the NPL market, and does not have a detrimental impact on EU NPL deals.
In line with that, the "mandatory" category should be much more stringent in scope and include only those data without which valuation on a loan-by-loan basis is simply not possible. It is advisable to concentrate on the core data necessary for trade.
For more detailed feedback, please refer to our answer to question 1 and see specific comments on the data fields in the dedicated columns of the data glossary (Annex II to the draft ITS).
What must be clearly stated in the document is the usage of the template – in which cases it is mandatory, and in which it is optional. (e.g., mandatory only in case when credit institutions decide to organize the sales NPLs trough the platforms, that require all mandatory information to be provided. In case of NPLs sales as a bilateral transaction, the seller and buyer may agree on the relevant information needed for due diligence and valuation of the NPL transaction.).
Moreover, in the part regarding the application of potential future supervisory activity must be amended “…competent authorities may, nevertheless, assess the availability of information and use of the template as part of their supervisory activities in the area of NPL management or credit risk management by the credit institutions, for cases when sale of NPL has been organized through the platforms.”
Namely, investors may accept loans with incomplete data, possibly at a lower price. Sometimes, the cost of collecting all mandatory fields is not justified by the potential expected price increase. We consider it particularly important that mandatory fields do not create an invincible obstacle for banks where some data is not available. In line with that, EBA must clarify what the consequences (if any) are for the bank as a seller in situations when it is not able to provide all mandatory fields.
Concerning the proposed set of templates, data glossary and instructions, these are not clearly defining how to `interpret` local legislation and product specifics into required structured data/fields. For example, there are several insolvency proceeding types across all of the EU; e.g. concerning rules and enforcement schemes applicable under local laws, this is still far from being harmonized even from a procedural point of view (how proceedings are initiated, what does this en-tail, etc.). The products/instruments themselves are not being standardized to give an equal information having some value in particular field entered either. Therefore, to standardise NPL transactions on secondary markets could be homogenous in terms of the product/enforcement view only on national levels. To achieve comparability, representability and standardization, there is more to elaborate on in the field of product standardization and harmonization of enforcement law and procedures across impacted countries/entities.
Some definitions still remain unclear or represent information that is not legally or reliably retrievable from the counterparty.
In this context, maybe there is a space for creating rating agencies for NPL loans (likewise for other assets) that would cooperate with banks to assess their instruments altogether with national enforceability specifications and such information with `several fields add-up` (on balance sum and its specification, collateral value and basic specs, arrears info, trigger for initiated court proceedings, last payment date and sum collected for last x months period or so...) would be a better option than trying to put different approaches/legislations/schemes into standardized set of values within particular fields – so letting the market decide what it really needs. If you asked NPL buyers whether they would like to have more and structured information, they would say "yes”, basically because of “just in case”. But if it comes to a pragmatic approach – no buyer would invest to gather so much data that does not directly say anything about debtor´s collectability (so why burden the banks?).
Instead, we propose to have a client (private individuals and legal entities) view and a product view (unsecured and secured) instead of having one absolute amount threshold. As a result, it is proposed to provide 23 fields in case of unsecured loans granted to Private individuals and an additional 10 fields in case of secured loans (highlighted as “Agreed” in the NPL template).
• CORP – floors / caps – single trx vs portfolio:
In our opinion, NPL data templates with significantly fewer data fields can be useful for portfolio sales, but not for single tickets in the (large) corporate segment or for commercial real estate financing. According to market standards for larger single exposures, sometimes no specific data is necessary for trades at all, because brokers and investors already have proper information, quotes indicate market price levels, sometimes buyers are already in a creditor`s position and increase their exposure. For large corporates the data template would not be sufficient, because more data is needed to assess the financial situation of a borrower, eg: business plans, going concern prognosis, independent expert business reviews, existing and future financial structure and creditors, existing and planned future shareholder structure, quality of management, …
Hence, the NPL data templates should NOT be mandatory for single ticket transactions (i.e. expo-sure of one group of connected clients) at all or at least there should be a cap defined and exclude single tickets with a nominal exposure exceeding a certain threshold, proposed € 3 Mio nominal value.
In general, we believe that for portfolios the proposed data fields are too numerous and not practical for transactions. Some defined data are not available in the banking systems and have to be collected manually, increasing cost, time schedule and reducing data quality.
The proposed threshold of € 25.000.- should be increased to € 300.000.- , to separate between retail business including residential mortgage on the one hand and the more complex corporate business on the other hand.
Regarding data governance, we are fine with the overall requirements related to completeness, consistency and accuracy of the data, but only for the data which we agree to keep in the NPL data, respectively marked “Agree” column P, Annex II
1. Do the respondents agree that these draft ITS fits for the purpose of the underlying directive?
ESBG disagrees with the proposed draft NPL templates for the following reasons:First, the number of mandatory data fields included in the proposed templates is significantly higher than what has historically been necessary to sign and close voluntary NPL deals, including secured and unsecured, residential and SMEs portfolios. This information is too detailed and is not deemed critical for purposes of loan valuations.
Despite a significant decrease of the data fields compared to the last template revision from May 2021, the remaining number is still high, and it contains notedly more information than the current market standard. In other words, it is still seen as too elaborate and granular. Some of the fields even surpass the information that is relevant for portfolio valuations – the general impression is that de-spite the revision and reduction of the previously named “critical fields”, data specifications are still very much statistically driven (evident by the compliance of many data points with values already known from AnaCredit /collected there, that are of a more statistical nature but do not affect the valuation or processing of exposures).
Also, as one of the participants in previous consultation sessions pointed out, there is a highly noticed influence of third parties with interests other than those of the sellers and buyers, as direct NPL transaction participants, which altered the nature of the templates more towards supervisory and sta-tistical purposes, as well as purposes of transaction platforms that have identified the data and its analysis and provision as a source of revenue, and advisors who want to sell their services around data validation.
Moreover, what has been noticed is that a lot of previously labeled “non-critical data” are shifted to category of “Mandatory” data with the latest draft, either without any definition change or with slight changes in field definition (increase from 70+ to over 130 mandatory data).
As a direct consequence of the increase in mandatory data fields, there would not only be a relevant increase in the credit institutions’ IT costs to include some of those new mandatory data fields in the banking systems, but also an increase in the time-to-market of the NPL portfolios, as some of the proposed mandatory fields may not be included in the systems within a reasonable time frame and the sellers would have to do it manually. This would probably mean an entry barrier to new potential entrants that have lower IT budgets.
Secondly, we think that it makes no sense to have common templates for single names or reduced portfolios of single names and massive portfolios of NPLs. Exposures to one single debtor or to a reduced number of corporates or SMEs have historically needed a different set of information, as their potential purchasers perform a deeper financial and legal analysis of the exposures rather than a statistical analysis, which is more adequate for massive portfolios.
Thirdly, the proposed templates would generate data inconsistency, as these would mainly contain management data, which would be incomparable between European banks and it would lead not only to misunderstandings among potential investors, but also to proposed unjustified price adjustments that may result in many transactions not being closed, due to comparing data between heterogeneous EU jurisdictions and practices.
Fourthly, the expected impact on the cost side for the sellers of NPL should be taken into consideration. Namely, there is generally a high correlation between high data quantity, potential for error and greater validation effort. This has been recognized by the EBA, and in the part referring to Internal governance, this added pre-sale validation burden is to be endured by the banks - the ITS draft requires credit institutions to establish an internal process ensuring that the information being provided to the prospective buyers has been validated by staff independent from the staff involved in the sales process and is subject to an appropriate managerial approval. This leads to a due diligence that should be carried out by the bank before it is even known whether a sale will take place (when demand and prices are still unknown) and building additional costs for the banks for data gathering and validation, causing the total transaction cost to increase. In other words, overly extensive data requirements create more complexity, effort, and liability risks – this in particular can create signifi-cant obstacles for smaller banks and might prevent them from entering the market, as they do not have sufficient NPLs in total numbers to amortize the additional cost for data validation.
Furthermore, the data sheet should have a disclaimer stating that the credit institution should not assume any liability for the information to be complete or sufficient for an in-vestment decision of the prospective buyers. The market for distressed assets on the buyer side is dominated by investment banks having specialized units trading and investing in distressed asset and by specialized distressed funds. The professional and analytical level on the buying side is very high. Investing in distressed assets is a high-risk business. We doubt that it would make sense to widen the investor basis in providing a data sheet as a basis for the necessary analysis. For an investment deci-sion the data will regularly not be sufficient. The analysis must be made by the buyer and its risk must be borne by the buyer as laid down, for example, in the LMA standards for distressed asset sales. If that risk partially moves to the seller side the market will not get bigger but significantly smaller. If data like the valuation of securities, or balance sheet information are included, the analysis partially moves to the seller side. Information that includes a valuation or analysis, in our opinion, should therefore not be part of the NPL Transaction Data Template. If certain information is provided in a data room, it should be possible to exclude this information from the data sheet. If for example the complete legal documentation is provided for the purpose of a legal due diligence by potential inves-tors, a data sheet summary of the legal structure should be obsolete.
Additionally, the aforementioned reasons imply that the proposal is misaligned with article 16(4) of the Directive, which states that, when preparing this draft ITS, both existing market practices in data sharing between buyers and sellers, as well as minimizing processing costs for credit institutions and credit purchasers, should be taken into consideration.
In conclusion, we would like to request that the EBA further streamlines the templates, aim-ing at simpler, more balanced and effective design in order to achieve a broader application and in-crease transparency in the NPL market, and does not have a detrimental impact on EU NPL deals.
In line with that, the "mandatory" category should be much more stringent in scope and include only those data without which valuation on a loan-by-loan basis is simply not possible. It is advisable to concentrate on the core data necessary for trade.
2. What are the respondents’ views on the content of Template 1? Please provide any specific comment you may have on the data fields in the dedicated columns of the data glossary (Annex II to the draft ITS) added for your feedback.
Please refer to our answer to question 1 and see specific comments on the data fields in the dedicated columns of the data glossary (Annex II to the draft ITS).3. What are the respondents’ views on the content of Template 3? Please provide any specific comment you may have on the data fields in the dedicated columns of the data glossary (Annex II to the draft ITS) added for your feedback.
Please refer to our answer to question 1 and see specific comments on the data fields in the dedicated columns of the data glossary (Annex II to the draft ITS).4. What are the respondents’ views on the content of Template 4? Please provide any specific comment you may have on the data fields in the dedicated columns of the data glossary (Annex II to the draft ITS) added for your feedback.
Please refer to our answer to question 1 and see specific comments on the data fields in the dedicated columns of the data glossary (Annex II to the draft ITS).5. What are the respondents’ views on the content of Template 5? Please provide any specific comment you may have on the data fields in the dedicated columns of the data glossary (Annex II to the draft ITS) added for your feedback?
This section combines both static and time series data, which will be highly problematic to provide in an efficient manner for a large number of positions in massive NPL portfolios.For more detailed feedback, please refer to our answer to question 1 and see specific comments on the data fields in the dedicated columns of the data glossary (Annex II to the draft ITS).
6. Do the respondents agree on the structure of Template 2 to represent the relationship across the templates? If not, do you have any other suggestion of structure?
Generally, we believe it’s properly structured. The contract identifier should however be the only requested information to represent the relationship across the templates, as this has been the market practice so far and it has worked fine for all parties involved in NPL transactions.7. Do the respondents agree on the structure and the content of the data glossary? Please provide any specific comment you may have on the data fields in the dedicated columns of the data glossary (Annex II to the draft ITS) added for your feedback.
See specific comments on the data fields in the dedicated columns of the data glossary (Annex II to the draft ITS).8. What are the respondents’ views on the content of instructions?
The instructions that are provided are rather detailed, but the quantity of requested data and argumentation supporting this quantity are not seen as fully supported and substantiated. We believe there is significant room for a further decrease of data fields in ITS, which will be more aligned with market standards and practices.What must be clearly stated in the document is the usage of the template – in which cases it is mandatory, and in which it is optional. (e.g., mandatory only in case when credit institutions decide to organize the sales NPLs trough the platforms, that require all mandatory information to be provided. In case of NPLs sales as a bilateral transaction, the seller and buyer may agree on the relevant information needed for due diligence and valuation of the NPL transaction.).
Moreover, in the part regarding the application of potential future supervisory activity must be amended “…competent authorities may, nevertheless, assess the availability of information and use of the template as part of their supervisory activities in the area of NPL management or credit risk management by the credit institutions, for cases when sale of NPL has been organized through the platforms.”
9. Do the respondents agree on the use of the ‘No data options’ as set out in the instructions?
In general, we see “No data option” as am improvement on the previous Consultation paper, however we noticed a significant impediment that needs to be taken into consideration referring to extension of applicability of all “No data option” (ND 1 – ND 4) to mandatory data.Namely, investors may accept loans with incomplete data, possibly at a lower price. Sometimes, the cost of collecting all mandatory fields is not justified by the potential expected price increase. We consider it particularly important that mandatory fields do not create an invincible obstacle for banks where some data is not available. In line with that, EBA must clarify what the consequences (if any) are for the bank as a seller in situations when it is not able to provide all mandatory fields.
10. What are respondents’ views on whether the proposed set of templates, data glossary and instructions are enough to achieve the data standardisation in the NPL transactions on secondary markets, or there may be a need for some further technical specifications or tools to support digital processing or efficient processing or use of technology (e.g., by means of the EBA Data Point Model or XBRL taxonomy)?
Please refer to our answer to question 1. Currently, there is already an active NPL market in the EU. The proposed templates will generate additional costs and obstacles that will not lead to improvements in the market, but to a slowdown as well as to entry barriers for small and medium sized sellers that might not be able to adapt their IT systems.Concerning the proposed set of templates, data glossary and instructions, these are not clearly defining how to `interpret` local legislation and product specifics into required structured data/fields. For example, there are several insolvency proceeding types across all of the EU; e.g. concerning rules and enforcement schemes applicable under local laws, this is still far from being harmonized even from a procedural point of view (how proceedings are initiated, what does this en-tail, etc.). The products/instruments themselves are not being standardized to give an equal information having some value in particular field entered either. Therefore, to standardise NPL transactions on secondary markets could be homogenous in terms of the product/enforcement view only on national levels. To achieve comparability, representability and standardization, there is more to elaborate on in the field of product standardization and harmonization of enforcement law and procedures across impacted countries/entities.
Some definitions still remain unclear or represent information that is not legally or reliably retrievable from the counterparty.
In this context, maybe there is a space for creating rating agencies for NPL loans (likewise for other assets) that would cooperate with banks to assess their instruments altogether with national enforceability specifications and such information with `several fields add-up` (on balance sum and its specification, collateral value and basic specs, arrears info, trigger for initiated court proceedings, last payment date and sum collected for last x months period or so...) would be a better option than trying to put different approaches/legislations/schemes into standardized set of values within particular fields – so letting the market decide what it really needs. If you asked NPL buyers whether they would like to have more and structured information, they would say "yes”, basically because of “just in case”. But if it comes to a pragmatic approach – no buyer would invest to gather so much data that does not directly say anything about debtor´s collectability (so why burden the banks?).
11. What are the respondents' views on the approach to the proportionality, including differentiating mandatory data fields around the threshold? Please provide any specific comment you may have on the data fields in the dedicated columns of the data glossary (Annex II to the draft ITS) added for your feedback.
In our view, the abovementioned excessive number of mandatory data fields is far from proportional, even after differentiating the threshold.12. Do the respondents agree with the proposed calibration of 25 000 euros threshold in line with AnaCredit Regulation? If not, what alternative threshold should be introduced, and why?
The higher the size of the position to be sold, the more information and documentation that might be available about it. To set the threshold at €25,000 would be counterproductive and have a negative impact in sales of massive NPL portfolios, especially in unsecured granular NPL portfolios.Instead, we propose to have a client (private individuals and legal entities) view and a product view (unsecured and secured) instead of having one absolute amount threshold. As a result, it is proposed to provide 23 fields in case of unsecured loans granted to Private individuals and an additional 10 fields in case of secured loans (highlighted as “Agreed” in the NPL template).
• CORP – floors / caps – single trx vs portfolio:
In our opinion, NPL data templates with significantly fewer data fields can be useful for portfolio sales, but not for single tickets in the (large) corporate segment or for commercial real estate financing. According to market standards for larger single exposures, sometimes no specific data is necessary for trades at all, because brokers and investors already have proper information, quotes indicate market price levels, sometimes buyers are already in a creditor`s position and increase their exposure. For large corporates the data template would not be sufficient, because more data is needed to assess the financial situation of a borrower, eg: business plans, going concern prognosis, independent expert business reviews, existing and future financial structure and creditors, existing and planned future shareholder structure, quality of management, …
Hence, the NPL data templates should NOT be mandatory for single ticket transactions (i.e. expo-sure of one group of connected clients) at all or at least there should be a cap defined and exclude single tickets with a nominal exposure exceeding a certain threshold, proposed € 3 Mio nominal value.
In general, we believe that for portfolios the proposed data fields are too numerous and not practical for transactions. Some defined data are not available in the banking systems and have to be collected manually, increasing cost, time schedule and reducing data quality.
The proposed threshold of € 25.000.- should be increased to € 300.000.- , to separate between retail business including residential mortgage on the one hand and the more complex corporate business on the other hand.
13. What are the respondents' views on the operational procedures, confidentiality and data governance requirements set out in the draft ITS?
We believe that the proposed operational procedures and data requirements are too complex and do not take into account the circumstances of all involved parties. Moreover, and in order to avoid breach of data protection rules, no personal data are shared with any potential sellers until the sale is closed with the intervention of a Notary (at least in Spain).Regarding data governance, we are fine with the overall requirements related to completeness, consistency and accuracy of the data, but only for the data which we agree to keep in the NPL data, respectively marked “Agree” column P, Annex II
Upload here the Annex II with your comments to the date fields
ESBG proposed NPLs template.xlsx
(930.62 KB)