As part of the Council Action Plan announced on 11 July 2017, the Commission, the European Central Bank (ECB) and the (European Banking Authority) EBA were invited to consider setting up NPL transaction platforms to stimulate the development of the secondary market. This paper provides considerations on this matter. The design features of the platform concept discussed herein draw on the article published in the ECB’s November 2017 Financial Stability Review 1 , and reflect further discussion and joint work by staff from the three institutions. This Staff Working Document is a product of collaboration between staff members from the European Commission, the European Central Bank (ECB), and the European Banking Authority.

This paper should not be reported as representing the views of the European Commission, the ECB or the EBA. The views expressed are those of Commission Services, the staff members from the ECB and the EBA.

Executive Summary

A European platform for Non-performing loans (NPLs) would be an electronic marketplace and a data warehouse where banks and investors could trade individual NPLs and NPL portfolios. Markets for non-performing loans remain poorly developed. A well-functioning transaction platform may have the potential to create active, liquid and efficient secondary markets for NPLs in Europe. It would contribute to reducing the degree of market failure in the European secondary markets for NPLs, helping to address information asymmetries, increase creditor coordination, and broaden the investor base, thus leading to improved market pricing of NPLs. As part of the Council action plan to tackle non-performing loans, the European Commission, the European Central Bank, and the European Banking Authority were invited to consider how such transaction platforms could be set up. This paper presents some practical considerations to this end, focusing on ownership, governance, operational requirements, and outlining selected legal issues.

In order to maximise the economic benefits and to provide a boost to secondary market for NPLs, a platform for NPLs should have a broad scope. In practice, this means that all loan types would be eligible for inclusion on the platform. The NPL platform would also be open to all types of sellers and professional buyers and its geographical scope of operation should be wide, preferably Union-wide. Secondly, in order to reduce transaction and search costs, the NPL platform would ensure data sharing and a high degree of data standardisation. To this end, an NPL platform could also, to the extent practically and legally feasible help improve NPL data quality by providing data completeness and plausibility checks, and defining standards for data validation that would be performed by banks. Moreover, it could facilitate transactions by offering a price discovery mechanism for participants to use and by intermediating between investors and third-party service providers such as appraisers, loan servicers, and transaction advisers. To avoid intractable legal issues, including issues related to insolvency and enforcement regimes, an NPL platform would never own, nor service any loans. Equally, settlement of transactions would be handled bilaterally between seller and buyer without involving the platform. In the medium term, the platform concept could be broadened to loans beyond NPLs 2

Currently, several private companies offer part of these services in a selection of EU Member States. However, they offer limited geographic scope and the loan data used is not standardised across the market. The supply of NPLs via existing platforms has been, until now, rather limited, in spite of significant investor interest. For these reasons, among others, the potential benefits of a European NPL platform remain largely unrealised.

Still, there is no clear-cut case for public ownership or operation of a European NPL platform for several reasons. Firstly, the operating costs of an NPL platform would likely dwarf the initial set-up costs and there is no indication that a public owner would be a more efficient operator. Secondly, national ownership of a European vehicle may be problematic both from a political and a ‘level playing field’ perspective. At European level, it is unclear what the appropriate ownership vehicle would be in view of the existing authorities’ regulatory and executive roles. Therefore, and on balance, it appears preferable to further build on the existing private initiatives. Keeping this in mind, an operating model based on private law initiatives could offer many advantages, including avoiding budgetary implications.

The role of Union institutions, in this context, would be to facilitate the identification by industry experts of proper operating and governance rules for European NPL platforms. These would be based on the principles outlined above and take inspiration from good market practices and existing Union law, where appropriate. An industry body would issue and enforce these industry rules. It would recognise platforms that meet them as European NPL platforms, and oversee compliance. In order to advance with setting up European NPL platform, a couple of steps could be taken. First, industry participants could be encouraged to start the work on setting-up an industry body needed to support the process of granting the ‘seal of approval’. Secondly, Union institutions and agencies, on one side, and Member States, on the other side, could discuss whether further and stronger incentives may be needed to spur participation by sellers, especially banks, in a European NPL platform.

Table of contents

1. Introduction

2. Current situation

2.1.Description of NPL markets

3. The concept of A European NPL Platform

3.1.Reminder: what is a European NPL platform?

3.2.General considerations for maximising the usefulness and efficiency of an NPL platform

4. NPL platform design features

4.1.Ownership and governance

4.2.Role of existing platforms

4.3.Role of Union institutions and agencies

5. Operational and functional requirements

5.1.Asset perimeter

5.2.Services to be provided by the platform

5.3.Use of NPL data templates and of data from existing public sources

6. Incentives and benefits to participants

7. Platform governance and legal considerations

7.1.Governance and control

7.2.Regulatory considerations

7.3.Data protection (data processing, data sharing, disclosure)

7.4.Possible issues arising from insolvency and recovery regimes

7.5.Scope of legal documentation needed

8. Conclusions and potential next steps

Annex 1: Examples of National NPL Platforms

1.Introduction

Within the context of the roadmap set out by the Commission on 6 December 2017 for deepening Europe’s Economic and Monetary Union (EMU) 3 , and in line with the Leaders’ Agenda 4 , one of the most immediate priorities is to complete the Banking Union and put in place all building blocks for a Capital Markets Union 5 . An integrated financial system will enhance the EMU’s resilience to adverse shocks by enabling private risk sharing across borders, while reducing the need for public risk sharing.

The Commission has also put forward other substantial measures in order to reduce risks and enhance resilience of the EU banking sector. For example, in November 2016, the Commission proposed a significant legislative package to review the Bank Recovery and Resolution Directive (BRRD), the Single Resolution Mechanism Regulation (SRMR), the Capital Requirements Directive (CRD) and Capital Requirements Regulation (CRR) with the objective of further reducing risks in the banking sector. 6 The same year, the Commission also adopted a proposal for a Directive on preventive restructuring frameworks, second chance and measures to increase the efficiency of restructuring, insolvency and discharge procedures. 7 Effective restructuring and insolvency rules are essential for the prevention or reduction of NPLs and for a well-functioning Capital Markets Union.

In order to further reduce risks, the Commission in March proposed a comprehensive package to accelerate the resolution of NPLs in Europe and to prevent their build-up in the future, together with its Second Progress Report on the Reduction of Non-Performing Loans (NPLs) in Europe. The package consists of two legislative proposals aimed to create a Union wide legal framework to encourage the resolution of NPLs and a staff working document providing non-binding technical guidance (a “blueprint”) for how national asset management companies (AMCs) can be set up. 8

In striving for the further development of secondary markets for NPLs, the Commission and other Union institutions and agencies have already taken a number of initiatives that could function as a catalyst for secondary market development. The initiatives, some mentioned above, are:

othe Commission’s proposed Directive on credit servicers, credit purchasers and the recovery of collateral 9 ;

othe Commission’s proposed amendment to the Capital Requirements Regulation regarding the introduction of ‘prudential backstops’ 10 ;

othe AMC blueprint;

othe EBA’s work on facilitating and improving data availability, comparability and quality (most notably the data templates first published in December 2017);

osupervisory guidelines on NPL management by the EBA and guidance from the SSM.

In this context, the Council action plan to tackle non-performing loans, adopted in July 2017, invited the EBA, the ECB and the Commission to consider the setting-up of NPL transaction platforms in order to stimulate the development of the secondary market. As a response, the ECB, in June 2018, presented work done in cooperation with the EBA and the Commission to the Council’s Financial Services Committee (FSC).

In eyeing a sustainable solution for the NPL problem in Europe, such a platform could provide a highly effective tool to prevent the renewed build-up of NPLs. As a platform could also aid in addressing existing NPL stocks, it would therefore reflect the essential objectives of the Council action plan. Furthermore, the usefulness of such a transaction platform could yield benefits reaching beyond NPLs. By increasing transparency of asset quality and reducing transaction costs, a well-functioning transaction platform may have the potential to create active, liquid and efficient secondary markets for NPLs in Europe.

The present document, which has been finalised in cooperation between staff members of the ECB, the EBA and the Commission services, gives further details on how to facilitate the emergence of NPL transaction platforms with a European scope. It recommends introducing a ‘seal of approval’ process for private NPL platforms. The document briefly recaps on the context within which platforms would be situated and explains in practical terms what the definition and purpose of a platform would be. It then discusses the important choices that would have to be made in the design of platforms. This is followed by consideration of important legal and governance aspects, including the platforms’ potential interaction with existing financial regulation and other relevant laws. Next, some practical aspects are introduced. The document concludes by considering the next steps that would be needed in order to advance further with the development of NPL platforms.

2.Current situation

2.1.Description of NPL markets

Despite some momentum in recent years, secondary markets for NPLs remain underdeveloped in Europe. While a genuine single market for NPLs in the EU may be difficult to realise in view of considerable cross-country differences in other relevant areas, in particular insolvency law, investors have been looking for opportunities across borders. 11

Markets for NPLs tend to be characterised by comparatively small trade volumes, a few large transactions involving a limited number of active investors, large bid-ask spreads when counterparts enter negotiations and a lack of transparency of market prices and volumes. 12 At the same time, Member States with higher loan sales recorded a stronger decline in their banks' NPL ratios. This suggests that secondary markets for NPL are contributing significantly to reducing NPL ratios.

Figure 1: Loan sales by underlying loan category

Source: COM calculations with KPMG data, retrieved from publicly available sources.

Between 2014 and 2017, transaction volumes in secondary markets for loans in the EU were estimated by industry sources to reach between EUR 100 billion and EUR 150 billion per annum 13 . The firms that collect data do not provide information about the share of NPLs in loan sales. The NPL market has also been highly concentrated on the buy side. The 10 largest transactions in 2015/16 accounted for one third of the transaction volume, while the rest was spread over about 480 transactions. Very few transactions were recorded with a volume below EUR 100 million 14 . Transactions have been strongly clustered in four countries: ES, IE, IT and the UK. In the first three, NPL sales have contributed substantially to reducing high NPL ratios. There have been few transactions in other countries with high NPL ratios (CY, EL, PT, RO, SI) and sizeable market activity in countries with low NPL ratios (UK, DE, NL). In Central and Eastern European countries, markets for NPLs seem to be emerging, but are still at an early phase of development 15 .

Of the 103 banks that disclosed transactions, about 40 had multiple transactions. The most important sellers were the National Asset Management Agency (NAMA) of Ireland and the Sociedad de Gestión de Activos procedentes de la Reestructuración Bancaria (SAREB) of Spain, which are national asset management companies created to acquire loans from Irish and Spanish banks respectively. The loan portfolios that banks sell include different asset classes and according to market sources, some buyers are specialised in specific asset classes. The figure below gives a snapshot of market shares by asset class based from a sample of 365 NPL transactions signed in 2015-2017.

Figure 2: Loans and NPLs on bank balance sheets per asset class and country – June 2018

Source: EBA

On the buy side, there are about 120 debt managers that invest in distressed debt in North America and Europe, of which about 70 are active in the EU 16 . In Europe, almost 40 % of the transaction deals was accounted for by the biggest five buyers. Most buyers are investment firms, but also a few banks bought loans, mostly from other banks 17 . About 70 % of the market share in the EU is controlled by 20 % of investors. The market is highly fragmented by national borders. The dominant number of investment firms was active in only one market and a few concentrated on two or three markets. The small number of investment firms active in four or more markets accounted for about a third of all transactions. Not all NPL buyers have been major investment funds, but they represent a sizeable market share. As regards the potential investor base for NPL, it is worthwhile to identify investment funds that specialise in comparable products.

Price data are usually not disclosed. Industry sources estimate that prices vary strongly depending on the type of debt and the quality of the underlying collateral (Figure 3).

Figure 3: Average price on face value of NPL portfolio transactions

Source: Association for Financial Markets in Europe (AFME) (2017) quoting PWC data.

Pricing is often an obstacle for transactions: if the prices offered are lower than the banks’ provision, banks make a loss, which reduces their capital and therefore inhibits their incentive to enter into a sales deal. These coverage ratios differ strongly across banks and Member States, in many cases being smaller in small banks than in large banks. They stand at around 46 % at the EU aggregate level 18 . Hence, for a price lower than 54 % (100 % — 46 %), the ‘average’ bank would have to record a loss. ECB staff estimated that the gap between book value and market prices for NPLs could be sizeable, amounting to between 11 and 45 % of gross book value (GBV), depending on the Member State. These estimates rely on the duration of recovery and recovery cost estimated by the World Bank and may not be fully representative of typical NPL transactions. Using a sample of about 30 transaction-level data points collected from a range of market sources, ECB staff nonetheless corroborated these theoretical estimates and found that there is little evidence that market prices of NPLs improved between mid-2016 and early 2018, despite a large body of anecdotal evidence suggesting that such improvement took place 19 .

3.The concept of A European NPL Platform

3.1.Reminder: what is a European NPL platform?

An NPL platform would be an electronic marketplace where holders of NPLs — banks and non-bank creditors — and interested professional investors can meet to exchange information and to trade. The scope of the platform could include other credit assets. The platform would function as a marketplace and would therefore not need to engage in trading on its own account or at any time own any of the loans traded. The broader the geographical scope of an NPL platform the more effective it would be in reaping the benefits of increased transparency and reduced transaction costs. Hence, EU-wide scope is preferable.

Figure 4: Potential functions for a platform

Source: ECB

In, theory, a platform should be set up to provide all of the following functions, depending on the degree of ambition, in order to maximise its potential benefits:

odata warehouse function, based on standardised data templates and qualitative information, and on provision of a seal of data quality to investors;

oa creditor coordination function, providing a forum for dialogue and decision-making by creditors with a claim on a common debtor;

oa transaction-facilitating function; and

opossibly, intermediation for further ancillary services, such as valuation and collateral appraisal, legal services, advisory services, and loan servicing, which would be offered by independent service providers to prospective buyers and sellers.

A platform has the potential to mitigate the impact of the market failures affecting the NPL market in a number of ways. It may improve and harmonise information and data availability. Moreover, it can reduce transaction costs associated with NPL transactions by reducing search costs and improving price formation. A platform can also potentially reduce market entry barriers, both by attracting a wider investor base and by making the platform particularly attractive to smaller banks that plan to sell NPLs. This will subsequently increase competition, which will lead to easier price discovery. Moreover, by making use of the standardised templates, a platform can further improve and harmonise information and data availability among market participants. Finally, it may help resolve coordination issues arising from multiple exposures (including exposure to the same borrowers). Importantly, it could provide a permanent channel through which future NPLs could be efficiently disposed of, as they arise. Crucially, it will also give impetus for the development of tools and services essential to secondary markets.

3.2.General considerations for maximising the usefulness and efficiency of an NPL platform

The main factor determining the extent to which the platform will be effective is the intake on both the supply and the demand side. In other words, the greater the number of sellers and investors using the platform, the greater the volume of assets offered and potentially traded on the platform, in a self-reinforcing cycle.

Currently, banks are not always able to manage their NPLs in an effective or efficient manner. In such cases, banks will recover less from their portfolio than a specialised party could achieve. This may occur, for example, when banks face a large volume of NPLs and are unable to properly manage their NPLs. Banks may also find themselves with a portfolio of legacy loans, for which their expertise in recovery may be of limited value. These may be, for example, loans that a bank has extended to borrowers outside of its core geographical market or loans backed by specific collateral types that the bank normally does not deal with. A bank may also have acquired non-core assets through a merger with or acquisition of another bank. In these instances, the best option may be to either outsource the servicing of these loans to a specialised loan servicer or sell the credit agreement.

It is therefore important that co-legislators continue to show resolve in finding an agreement on the proposed Directive on credit servicers, credit purchasers and the recovery of collateral. That will contribute to the development of secondary markets for NPLs by removing undue impediments to credit servicing by third parties and to the transfer of loans to credit purchasers. The proposed Directive, which fully respects the existing Union civil law acquis and Member States’ consumer protection rules, creates a common set of rules that credit servicers need to abide by to operate cross-border within the Union. The proposal sets common standards to ensure proper conduct by and supervision of credit purchasers and servicers across the Union, while allowing more competition by harmonising the rules on market access across Member States.

This will lower the cost of entry for potential credit purchasers by increasing the accessibility of credit servicing and by reducing the costs of credit servicing. A higher number of purchasers on the market means a more competitive market, leading to higher demand and transaction prices. Each of these elements will increase the volume of both NPL transactions and bid prices for NPLs. An effective ecosystem of third-party service providers may also be important in attracting smaller sellers and investors. On the supply side, banks that hold a significant amount of NPLs should be incentivised to dispose of NPLs so as to make their balance sheets more resilient, and ultimately to improve lending to the economy and to reduce risks to financial stability.

The availability of high quality loan-level data is a prerequisite for the successful and efficient functioning of secondary markets and a key advantage of NPL platforms. NPL platforms, by using standard data definitions and templates, will improve the quality and comparability of data available to market participants. With the wider use of EBA NPL data templates, where the platform provides an impetus, banks are expected to make considerable efforts to further improve their data.

At national level, Member States could usefully be encouraged, perhaps in the context of the European Semester, to foster participation in the growth of European secondary markets for NPLs and to undertake necessary reforms towards remedying the problem of NPL non-transferability and other challenges to NPL work-out, including costs and time to enforcement and predictability of enforcement processes. Already, issues related to NPLs are closely considered in the European Semester. They have been analysed in several country reports and several country-specific recommendations (CSRs) on these topics have been adopted.

4.NPL platform design features

4.1.Ownership and governance

Given the light footprint of an NPL platform and the comparatively low cost of setting it up, there do not appear to be significant barriers to setting up a multi-jurisdiction or even EUwide transaction platform. A wider range of loans could be available on a platform with EU-wide scope, costs could be shared and reduced, the scope of assets could be broadened beyond the domestic sector, and services from across countries could be provided through the platform.

A European platform covering multiple Member States would therefore have an advantage over purely national platforms. A European platform would be able provide a consistent, standard service to large investors, who operate in several Member States, while remaining open to smaller investors who often are active only in a single country. However, its cross-border scope would only increase efficiency as long as data and access procedures are harmonised across national borders. Additionally, the provision of ancillary services at cross-border scale may be limited by the fact that markets for such services are highly fragmented and the platform would need have a local presence (either through local partners or in the individual jurisdictions it would cover).

The NPL platform would be a marketplace and have some features of a market infrastructure. There is no single established best practice in designing ownership of marketplaces or market infrastructure. Some critical market infrastructure providers are publicly owned (e.g. payment systems such as TARGET2), yet others are for-profit private companies (e.g. many stock exchanges) or industry initiatives operating in a not-for-profit (European Data Warehouse) or cooperative form (SWIFT) 20 . Several forms of ownership and governance can be considered for an NPL platform, including (1) public ownership, (2) private ownership, (3) private ownership combined with standard setting and oversight by a not-for-profit ‘standard setting body’.

Public ownership of an NPL platform may mean ownership by a specific Member State or by a Union body. Cross-border operations of a platform owned by a specific national government would be highly sensitive from a political and communications perspective, and might not be viable. National ownership might also pose challenges under the EU State aid framework.

Pan-European public ownership would fit better with the European scope of the platform and avoid most potential State-aid issues. However, it would also pose some challenges:

·Selecting an appropriate Union institution or agency as the owner of the NPL platform. An appropriate owner should ensure that the platform would be free from political interference. On this count, while pan-European ownership is likely to reduce politicisation of the platform, safeguards would still likely be needed to prevent a block of Member States from influencing the strategy and business decisions.

·Finding solutions to effectively managing the potential conflicts of interest the Union institutions and agencies in their various regulatory and executive roles, and the business of the platform.

·Finding ways of ensuring that a publicly owned platform would achieve the same operational efficiency as a private enterprise could do, given the constraints imposed on operations of the public sector.

The recurrent operating costs of an NPL platform will likely dwarf the initial set-up costs and there are no indications that a public owner would be a more efficient platform operator than a private one. Therefore, in the case of public ownership, on-going public support covering platform operations could be needed. Consequently, public support covering the initial setup of a platform only, would likely not be a sufficient solution.

Public ownership could be justified if the respective entity enjoys a natural monopoly. An NPL platform would definitely be a case of increasing returns on scale, and possibly on scope. Barriers to entry for a new competitor would be also rather high if a platform is well established in the market. However, it is not evident that it would be a natural monopoly.

Hence, there is no clear-cut case for public ownership and should appropriate private-sector providers be found in the market, it may be more appropriate to support the growth of these private initiatives. Insofar as the platform may be likened to market infrastructures, there is no consensus on what operating and ownership model would be most appropriate. Some of the existing market infrastructures operate as for-profit companies (e.g. many stock and derivatives exchanges) and are owned by financial investors. However, market infrastructures may also be user-owned.

For these reasons, and in order to maximise the benefits of a European NPL platform, the governance model could usefully be one of privately owned NPL platforms that would meet the criteria identified in this document for improving the functioning of secondary markets for NPLs. In order to ensure this, the private platforms would seek a ‘seal of approval’ by an industry body tasked with developing industry standards and assessing if NPL platforms meet these criteria. In exchange, a compliant platform could be recognised as a European NPL platform. The industry body would also oversee the continued adherence of the platform(s) to these standards.

The governance, organisation and operating model of the NPL platform could therefore be based on industry standards regarding data quality, which a European NPL platform should follow. Moreover, a platform rulebook covering good governance practice and ensuring that a recognised European platform offered its services across the EU would help increase the transparency of secondary markets for NPLs. Further details on the necessary documentation are offered below in Section 7.5.

In this regard, some parallels can be drawn to the operation of accounting standard-setters (e.g. IASB) 21 and to the Asset Backed Securities Loan-level Data initiative. 22 Moreover, in the setup of such a structure, inspiration could also be taken from the operations of the International Swap and Derivatives Association (ISDA) 23 in derivatives markets as well as the Global Legal Identifier Foundation (GLEIF) 24 .

The industry body that would define the requirements for European NPL platforms could bring together representatives of the main stakeholders: banks, NPL buyers, advisers, providers of ancillary services etc. Union institutions and agencies could in line with the European scope of the platform and scope of the platforms participate as observers. 25 The body would scrutinise the operations of individual platforms, and provide a ‘seal of approval subject to compliance with the data standards and other requirements. From a legal perspective, this non-profit body setting industry standards for European NPL platforms could be established as a foundation, or similar. However, other forms of incorporation may be found to be suitable candidates, as well.

The ‘seal of approval’ process may lead to multiple platforms operating in the marketplace. As long as these platforms meet the required standards and functional requirements, including with respect to the services offered and geographic scope of operations, there is no case for excluding one or more. However, to avoid multiple exchanges segmenting the market, there would be great advantages to be had from linking the data warehouses operated by these platforms, so that an investor could access the market through a single point of entry. 26

4.2.Role of existing platforms

Some players active in several Member States offer services that are in part similar to the scope of the NPL transaction platform as discussed here. Some of these players operate on a cross-border basis, Union and international, while some others are present only in individual Member States. There are also some platforms where Member States have played a role in setting them up. These also have various shapes, functions, and ownership structure. In general, though, they are limited to loans from one Member State or a selection of banks within that Member State. Further details can be found in the annex to this staff working document.

None of the existing players offers the full scope of services that the NPL platform outlined here should offer. Some players provide loan data warehouse and auction platforms, including data analytics and valuation services, while others are less comprehensive and focus only on creditor coordination or specific market segments. In spite of the fragmentation of and limited offerings available in the marketplace, these firms may be able to expand, at a lower cost than a potential start-up, to the desired scope of the NPL transaction platform. Solutions available in the market appear easily scalable across borders and already cover a wide range of asset classes.

In view of the existing capacity, which in particular in the field of database and marketplace solutions appears highly advanced, it may be desirable to make use of existing platforms when fostering the growth of a European NPL transaction platform. The existing providers could seek a ‘seal of approval’, that would identify players which would be then committed to meeting the requirements set out in the prospective specifications of a European transaction platform on an ongoing basis.

Market intelligence suggests that all existing platforms are hampered by the limited supply of NPLs. However, they do not appear to face problems in finding interested investors. This, in turn, points to an incentive issue from a seller’s perspective that may be holding back growth of such private initiatives.

While there are clearly synergies and economies of scale and scope at play, which may leave limited room for more than one platform in the EU, the ‘seal of approval’ process would not restrict the number of platforms. Existing platforms, therefore, could continue to operate, albeit subject to evolving to meet the centralised standards.

4.3.Role of Union institutions and agencies

The Union institutions and agencies, in particular, the Commission, EBA and ECB, could play a range of roles in fostering the implementation of the concept of the NPL platform:

·In their law-making and regulatory role, they could prepare relevant regulations governing the operations of NPL platforms, if necessary. They could also assist in the review of existing regulations and propose removal of obstacles to this concept, again if necessary.

·Union institutions and agencies could contribute to shaping the desired requirements for a qualifying NPL transaction platform and participate as observers in the standard-setting body, which would provide a ‘seal of approval’ to NPL platforms and allow them to make full use of the incentives for the sellers

·They could help coordinate national efforts to prepare the banking sector for the transition to the NPL transaction platform, insofar as there would be new Union law requirements that need to be met.

·Critically, Union institutions would identify and propose to introduce incentives that help overcome the lack of supply of distressed assets.

·The Union institutions and agencies would, finally, encourage national authorities, in particular with respect to addressing remaining national obstacles to the development of NPL markets (e.g. through country-specific recommendations where appropriate).

·Should a fully private platform not emerge despite efforts made, other ways to facilitate the expansion of NPL platforms could be considered.

5.Operational and functional requirements

5.1.Asset perimeter

The success of the NPL transaction platform would depend largely on its capability to reach a sufficient (critical) size and of realising the economies of scale and scope. It therefore seems warranted seeking the greatest possible size for the platform in terms of the value of assets offered for sale.

For this purpose, the NPL transaction platform should ideally host a broad range of asset classes 27 . For ease of access and comparison, assets should be grouped into more homogenous groups, such as:

·commercial real estate (CRE) assets,

·residential real estate (RRE) assets,

·SME or other corporate loans,

·unsecured retail loans,

·asset-backed finance,

·car loans,

·other/specialised NPLs.

Against the backdrop of the expected benefits of the platform, particular emphasis could be put on those loans and portfolios where a single borrower may be indebted to several banks (i.e. multiple lender relationships), and where coordination issues could arise. These, most likely, are corporate, SME, commercial real estate, and specialised lending portfolios. Besides NPLs, a platform could facilitate trading in performing loans and loan portfolios that are offered for sale. Offering performing and non-performing loans on a single platform is fully compatible from a legal, technical and reputational point of view.

The scope of the platform does not need to be fixed during the set-up phase but can be left flexible. No minimum size threshold would be required for loans and/or loan portfolios as the market will determine, if needed, a threshold below which it is not viable to maintain small or highly granular exposures.

While all loan types would be accepted by a European NPL platform, it might attract some portfolios more easily. For example, due to cost efficiency, larger corporate portfolios could be expected to be attracted in the first instance. Another important advantage of the platform as regards large corporate portfolios is that it could help overcome coordination challenges, as large corporates usually have outstanding debt to multiple creditors 28 . Corporate loans are easier to handle than consumer or mortgage loans given the application of personal data and consumer protection rules to the latter. This means that a platform would need to fully comply with all existing personal data and consumer protection rules, including the General Data Protection Regulation (GDPR) and the e-Privacy Directive 29 . That said, consumer or mortgage loans are more standardised, hence potentially better suited to being traded centrally on an NPL platform than for example SME loans.

5.2.Services to be provided by the platform

In order to be eligible for a ‘seal of approval’ the platform should directly offer at least the following services across EU Member States:

Data review and validation: ideally, a platform would provide data that are quality-assured and subject to a range of checks, as this would be one of its key contributions and success factors. However, full data validation would have to rely on available third-party expertise which could be costly, and, at least in the early stage of developments, prohibitively so. Moreover, data validation may be constrained by personal data protection rules.

·In the initial stage of development, the platform may accept that data are not fully validated. Nonetheless, it should implement some automated data quality checks and validation rules to ensure at least plausibility and a certain level of completeness of the data. It could identify critically relevant data fields and focus on those being populated. Information on these checks should be accessible to investors. Importantly, the NPL platform should also enforce the use of common identifiers to allow investors to identify common borrowers in datasets coming from multiple banks. More stringent data validation could remain, at least initially, voluntary and would be carried out at the discretion of the NPL seller who then should provide an overview of applied checks and their methodology. However, even at this early stage, a minimum data validation standard should be established by the standard-setting body, which should encourage qualifying platforms to adhere to that standard when preparing datasets.

·In the steady state, the platform should accept only fully validated data. External data validation should use tools and techniques that are in line with the standards defined within the scope of the ‘seal of approval’, so that the investors would be familiar with the data checks and could take comfort from the validation reports. Owing to the volume and predictability of the data validation needs, the platform may be able to offer cost savings by contracting data validation in bulk with respect to the current, decentralised model where every investor engages its own consultant or internal experts. The expense should be directly covered by the NPL sellers, as they are responsible for ensuring the data quality of the portfolio in the first place and will respond to a price incentive in this regard.

Data warehousing: the platform would operate an electronic database, regularly updated with new snapshots of NPL loan-level data, and provide access to investors. 30

Matching buyers and sellers of NPLs: as a key function, the platform should allow investors to contact the NPL sellers and to bid for selected NPLs or NPL portfolios. The specific details of the price discovery process remain to be determined. In addition to straightforward matching of buyers interested in purchasing whole portfolios from single sellers, the platform could also offer matching and bundling of (small) similar loans that specific investors might be interested in purchasing, as opposed to buying the whole portfolios put up for sale by the banks.

Other, ancillary services to be possibly provided by third parties and offered through the platform may include the following:

·Intermediation of credit servicing: while some investors cooperate with an existing servicer, or use a captive servicer, to help manage acquired NPL portfolios, new investors may not have an established servicer relationship. They could benefit from being able to find and choose an appropriate servicer when contacting the platform. Many credit servicers would need to be close to the operations of the platform, for example as data providers (e.g. for portfolios serviced on behalf of banks). Therefore, there would be limited additional effort needed to allow servicer selection via the platform. Credit servicers mainly serving loans on behalf of buyers would have a natural incentive to be associated with such an initiative.

·Valuation: a broad range of analytics and valuation services could be provided by third parties. On the platform, the internal IT system could offer some standard quantitative tools, for loans and for the underlying collaterals, and data that can be used for benchmarking purposes based on similar past transactions. This could be coupled with more bespoke advisory services given by third parties. These could include corporate valuation, individual real estate appraisal, etc. In these cases, the platform could act as an intermediary, similar to the case with servicers.

The transaction platform could also be developed in stages to ensure that the concept can be launched quickly. Alongside an introduction of the data validation function in stages, the ancillary services may be phased in after the basic functions of the platforms are operational. Furthermore, and to the extent that it follows existing rules on giving investment advice, the platform could also offer various data analytics tools (either directly or through third-party providers) helping sellers to arrive at their own valuation and hence to determine potential asking prices, or access to pre-selected deal advisers that could help buyers to prepare sales proposals and data, and to identify potentially interested investors.

5.3.Use of NPL data templates and of data from existing public sources

EBA NPL data templates are highly granular loan-level templates. They cover all the data that investors might need to analyse and perform proper and detailed financial analysis of an NPL portfolio. These include detailed information about the loan characteristics, counterparties, collateral and status of the enforcement process. They cover all loan types (including residential, CRE, SME/corporate, unsecured, leasing / asset-backed finance (ABF), auto, others/specialised). Each data field in the templates is classified by relevance to financial due diligence (FDD) and valuation purposes, ranked as ‘critical’, ‘important’ or ‘moderate’.

However, not all fields are needed for the initial portfolio screening and the early valuation stage by a prospective buyer. This suits the role to be carried out by European NPL platforms, which is to disseminate portfolio information to potential investors that is only fit for the initial phases of due diligence and valuation of NPL transactions. Hence, transaction platforms should use a subset of critical data fields, as pre-defined in the EBA NPL data templates. Sellers, transacting through the platform, should be required to follow an established minimum comprising the ‘critical’ data fields in order to be able to process any transaction through the platform. This would enable investors to efficiently and accurately value loan portfolios. In the later stages of due diligence, for example when information is made available through the virtual data rooms, platforms should facilitate the exchange of the complete dataset of EBA NPL data templates. The common usage of the standardised templates will facilitate loan-level comparison and application of various data analytics to be offered by or through the platform. The EBA NPL templates by design can enhance granularity, quality and comparability of NPL data. Therefore, the NPL transaction platform(s) should use the EBA NPL templates to facilitate initial portfolio screening and FDD and valuation by prospective buyers.

As explained in section 3.5, data validation is a key contribution and success factor of this secondary market initiative. The platform(s) should offer minimum data quality services or data validation to ensure that the data provided is complete, accurate and consistent. The data quality could be checked at the following levels.

First level:

·Completeness: all required data fields are reported;

·Consistence and coherence: data abides by simple automatic validation rules.

Second level:

·Plausibility: this might be checked by the platform in order to raise flags on an unexpected result but would not automatically signal an error.

Third level:

·Accuracy: this might entail random checks also involving ancillary services (e.g. servicers or advisers) to verify the accuracy of data.

Only the first two types of data quality checks can be easily automated within the IT systems of the platform. For this purpose, platforms could build on the data validation rules included in the EBA NPL templates. It is important that the platform utilise other possible sources of market data (e.g. real estate data for collateral valuation) and building on their expertise in providing comprehensive data quality services. Data accuracy checks, however, are unlikely to be possible within the platform, as they require labour- and time-intensive interactions with the seller and access to loan-level files.

As explained above, it makes sense that NPL platforms require a subset of ‘critical’ fields, as defined in the EBA NPL data templates, which would enable buyers to efficiently and accurately value the portfolios for sale and submit an indicative price to the seller. This subset should include those fields with the highest relevance to financial due diligence and valuation purposes. Although the standardisation of these templates is of the utmost importance, some level of flexibility will be needed for sellers regarding the number of data fields that are mandatory. Therefore, NPL platforms should not accept portfolios with more than 15 % of the data fields considered ‘critical’ submitted empty.

However, for some legacy portfolios, it may make little economic sense for the seller to invest in resources to provide a higher quality of data to prospective buyers. This would primarily concern loans with limited expected recoverable value. Exemptions could be considered for such portfolios, which are expected to decrease in importance as banks improve their internal controls and loan monitoring.

6.Incentives and benefits to participants

Although the incentives are applicable to an accredited platform, many of the incentives are general and are inspired by well-functioning NPL secondary markets. For example, using a rich and consistent data set for the transactions on the NPL platform would allow the platform (and possible third-party data analysis providers) to provide better deal and price statistics to all parties in the process. It would also allow the running of more precise valuation models giving sellers the ability to set realistic asking prices, and giving buyers more information on the value of the loans in question.

Finally, other economic factors could motivate market participants (investors/buyers and institutions/sellers) to use the platform such as economies of scale due to lower transaction cost, price discovery and transparency, and higher market confidence due to the ‘seal of approval’ and attached platform governance requirements.

Nevertheless, transactions through an accredited NPL platform can bring substantial incentives for all market participants, as set out below.

Potential incentives and benefits for investors/buyers transacting through an accredited NPL platform

·The platform will improve and harmonise information/data available to buyers by making use of harmonised data templates, based on EBA NPL templates. Investors can use such validated and harmonised data in the financial due diligence and valuation process, and they can accurately perform portfolio valuations with significantly reduced costs. Using an NPL platform will provide an incentive to potential investors, as they would know in advance the standardised data templates provided, ensuring comparability and consistency in their valuation models. The data validation offered on behalf of the platforms would also facilitate the process and accuracy in the valuation.

·Through its data analysis capabilities, the platform would help investors to obtain information on the specific portfolios/sectors/borrowers across different sellers. This would allow buyers to target specific market segments or borrowers in their purchase strategies and adjust their existing portfolios of loans. To facilitate this, the NPL platform could proactively offer matching and bundling of (small) similar loans that specific investors might be interested in purchasing (as opposed to buying the whole portfolios put for the sale by the banks). This could create critical mass and therefore portfolios that are more attractive to potential investors. This could expand the reach of buyers to NPL portfolios held by smaller institutions and to other geographical markets.

·Through the ‘seal of approval’ procedure and the use of standard and clear rules and procedures for sellers and buyers via legal framework and standardised data infrastructure, the platform would increase market transparency, which would also act as an incentive for investors.

·In addition, the platform would be able to provide potential investors with the contacts of (independent) servicers or may also affiliate with various servicers in different Member States in order to provide investors with direct access to these related services. Potential investors, especially those without access to own servicers (through either affiliation or partnership) would have more options and would not necessarily have to look for independent servicers or rely on services provided by the sellers. This is increasingly important for investors when they are expanding into markets ‘unknown’ to them.

·Similarly to the provision of access to servicers, the platform could also act as an intermediary for potential investors and give access to selected law firms and valuation experts in the Member States in which it operates. That would reduce the potential entry barriers for the investors new to some markets.

Potential incentives and benefits for sellers transacting through an accredited NPL platform

·Through its user agreements, the NPL platform would have a database of investors interested in the purchase of various types of NPL (and other assets at a later stage) generating a pool of investors that banks could tap in one go, instead of approaching investors directly.

·One of the prerequisites and benefits of using the platform is a standardised approach to the portfolio and transactional data based on the EBA NPL templates. Incentivising banks to use the data templates at the point of NPL sales through the platform would also ensure that the banks have the required data in-house, ideally starting already from the point of the origination of a loan. The EBA could facilitate this through its draft Guidelines on loan origination and monitoring (currently under development), where it could introduce a requirement for banks to consider EBA NPL templates as an appropriate way of structuring internal data for internal monitoring of loans from the point of their origination. In this way, all three initiatives (the platform, templates and Guidelines on loan origination) would be linked 31 . Such monitoring activities will be assessed by the competent authorities on a continuous basis as part of the supervisory review and evaluation process (SREP), and banks not meeting the requirements may be subject to the supervisory measures, as shortcomings will be considered as internal governance or control weaknesses. From the supervisory side, competent authorities could recognise the use of the NPL transaction platform as an advantage in their SREP assessments of the institutions’ NPL strategies and overall credit risk management and controls, and could reflect (positively) in supervisory scoring and overall assessment.

·The platform would offer data quality checks and data validation for the purposes of entering new transactions, and this would help banks with providing data necessary to investors. As part of data validation, in addition to automatic validation rules the NPL platform can offer additional data services, for example a data help desk to assist banks on building the NPL standardised data sets. The latter could be done either by the platform itself, or through third parties (e.g. services) specialising in the NPL from certain Member States.

·Depending on the relevant legislation 32 there could be benefits for institutions which use the platform and the standardised templates (e.g. to meet data requirements for a transfer or for any centralised NPL registry or other supervisory data requirements).

Sellers, especially small ones, could benefit from the platform, as it would allow them to target a wider range of potential investors.

Potential incentives and benefits for service providers stemming from an accredited NPL platform

·Specifying a ‘seal of approval’ for NPL platforms will create incentives for various ancillary service providers, in particular loan servicers, data analytics providers, deal advisers and law firms, to affiliate with the platform and offer their services to potential investors who otherwise do not have access to their own service providers.

The EU-wide reach of an accredited NPL platform would allow existing service providers that operate in narrower segments (e.g. loan servicers specialising in specific countries) to expand their offering and cover more markets. At the same time, however, the requirement for the platform to have EU-wide scope may let the NPL platform seek partnerships with the existing EU-wide service providers (or at least firms providing services in multiple Member States), and thus disadvantage smaller players. Also, on the deal advisory side, introducing requirements for the platform to grant access to recommended advisers could incentivise consulting and advisory firms to engage with the platform.

7.Platform governance and legal considerations

7.1.Governance and control

The level of governance and control that should be required from the platform would depend on the services offered by the platform and its operations. The platform, as described in this note, would neither provide settlement services, nor at any time assume ownership of the assets transacted or face financial risks related to the value of the loans transacted. Moreover, it would not be subject to counterparty risk stemming from either buyer or seller. Finally, the seller would retain all legal responsibilities vis a vis the borrowers until they are transferred to the prospective buyer.

Consequently, there are no strong reasons to require formal authorisation of the platform as an exchange or other type of regulated marketplace as that would likely trigger an excessive level of regulatory scrutiny and associated compliance requirements. If this option were considered at a later stage, however, any form of direct supervision of the platform would likely require changes in primary EU legislation, with clear allocation of supervisory responsibilities, and incorporation of prudential requirements.

In the absence of direct supervisory requirements, any platform would be subject to the corporate law and corporate governance requirements in the Member State where it is incorporated. Depending on the services provided and the Member State, the platform may also fall under the scope of national market conduct legislation and/or supervision.

Furthermore, an accredited platform would fall under the governance requirements identified by the industry body granting the ‘seal of approval’. If deemed necessary these could be developed by extending existing sector-wide governance (and outsourcing requirements) to the NPL platform. This could include inter alia fit and proper assessments of platform members’ management body using existing standards for internal governance (e.g. EBA Guidelines on internal governance, and Joint ESMA and EBA Guidelines on the assessment of the suitability of members of the management body). Such requirements can form the basis for the governance requirements in the documentation of the standard-setting body. Similar extension of the EU governance requirements could be considered in the field of market conduct requirements; the EU requirements could be incorporated through the rulebook for European NPL platforms. Importantly, the operational risk associated with the platform’s operations, due to either business or event risk, will need to be considered in the governance requirements set out for accredited platforms.

7.2.Regulatory considerations

There are a number of EU regulations and directives (existing as well as pending) that could be applicable to an NPL transaction platform. This section provides a brief description of the main interactions with applicable law.

Markets in Financial Instruments Directive (MiFID II) and Markets in Financial Instruments Regulation (MiFIR) 33

Since a platform could comprise a multi-country electronic marketplace where holders of NPLs — banks or non-bank creditors — and accredited investors have an opportunity to trade, i.e. matching buying and selling interest. The MiFID II/MiFIR framework could, in principle, have a bearing on the design of the NPL transaction platforms. This is especially the case insofar the as respective services and activities fall within the scope of Section A of Annex I of MiFID II. Against this background, the private law standard-setting body discussed above could take into account, among others, the following considerations from MiFID II and MiFIR when drawing up its accreditation requirements:

oorganisational requirements

omarket transparency, trade data, disclosures

ointegrity

oconflict of interest rules (if applicable)

oaccess rules

osupervision, competences, sanctions.

Market Abuse Regulation 34

The EU market integrity legislative framework prohibits ‘insider dealing’, ‘market manipulation’ and ‘unlawful disclosure of inside information’. Under certain conditions, the circumstances of a specific NPL-related transaction — say, a successful sale of a large portfolio — may amount to using ‘inside information’. This is defined as information of a precise nature, which has not been made public, relating, directly or indirectly, to an issuer or its financial instruments, and which, if it were made public, would be likely to have a significant effect on the prices of those financial instruments or on the price of related derivative financial instruments. Under such circumstances, the following considerations are apposite: transparency requirements (disclosing the specific piece of inside information vs delaying the disclosure), certain safeguards and prohibitions (e.g. prohibition on insider dealing), public disclosures, insider lists, managers’ transactions, etc.

Alternative Investment Fund Managers Directive 35 .

Platforms would be open to professional investors only entailing coverage by the Directive and there is no obstacle to fund managers’ buying or selling NPLs via such a platform.

Banking legislation

The Capital Requirements Regulation and the Capital Requirements Directive do not provide specific rules for selling/buying loan (portfolios) or other assets via transaction platforms, i.e. ‘standard’ governance and management requirements and requirements for credit, market and operational risk apply 36 . In this context, operational risks resulting from the use of transaction platforms appear most relevant. EU legislation requires that institutions adequately manage and mitigate operational risk, which is defined as the risk of losses stemming from inadequate or failed internal processes, people and systems or from external events. Operational risk includes legal risks but excludes reputational risk and is embedded in all banking products and activities.

Depending on the concrete set-up and service arrangement with the platform, outsourcing requirements could in theory apply. However, institutions would not rely on the transaction platform for data warehousing or other services that would otherwise be undertaken by the institution and would be subject to the forthcoming EBA guidelines on outsourcing arrangements.

Securitisation framework

While the main purpose of the platform would not be to facilitate securitisation of NPLs, or to trade (performing) securitised loans, the data provided via the platform may be very helpful in securitising loans. If securitised loans would be considered eligible for inclusion on the platform, the securitisation framework (the Regulation covering simple, transparent and standardised (STS) securitisation) would apply 37 . The products would, in order to attain classification as STS, need to comply with the requirements of the Securitisation Regulation, namely:

odue diligence requirements for institutional investors;

orisk retention by the issuer (at least 5 %);

otransparency requirements for the issuer, in particular disclosing information on the underlying exposures on quarterly basis;

ocredit-granting criteria — verifying that the underlying exposures were originated following the same credit-granting criteria as the non-securitised exposures;

oNPL securitisation would require disclosure of information to securitisation repositories as well. Therefore, it may be desirable that information disclosed on the NPL transaction platform follows closely the established format for reporting to securitisation repositories.

However, it should be noted that a securitisation of NPLs cannot be considered STS.Interaction with civil law and procedure rules:

In this context, the limited scope of services of the platform discussed in this document should be noted. The interaction with private international law would need to be further analysed in the definition of industry standards as well as in relation to each individual platform. Potential NPL sellers and potential NPL buyers would, independently of the bilateral NPL purchase agreement concluded between them, also enter into contractual relationships with the platform for the use of its services, determining their respective rights and obligations already in the pre-acquisition, due diligence stage. Any impact on applicable law and jurisdiction would have to be determined in each case.

Proposed Directive on credit servicers, credit purchasers and the recovery of collateral 38

While setting up the standards for transaction platforms, the following provisions concerning the Commission proposal for a Directive on credit servicers, credit purchasers and the recovery of collateral should be duly borne in mind:

·Eligibility criteria for bidders and credit servicers: The proposal does not set any authorisation criteria for credit purchasers and thus there appear to be no constraints regarding the eligibility criteria that a transaction platform might wish to set for them. In case the transaction platform entails tasks for credit servicers that are neither a credit institution nor a subsidiary of a credit institution, the authorisation framework for credit servicers established with the Commission proposal will be relevant.

·Requirements for third-country credit purchasers: The proposal states that (i) they would have to designate a representative who is domiciled or established in the Union; and (ii) for portfolios covering credit agreements concluded with consumers, that they would have to appoint a credit institution established in the Union or an authorised credit servicer to perform the credit servicing activities.

·Information obligations: The proposal would require creditors to provide all necessary information to the credit purchasers and mandates the EBA to develop technical standards for the formats to be used by creditors for the provision of detailed information to credit purchasers, for the screening, financial due diligence and valuation of the credit agreement. These could also be relevant for the transaction platform. The proposal would also impose specific obligations to inform competent authorities about credit transfers in order to enable them to effectively follow up on any interference of credit purchasers or credit servicers with borrower protection rights.

·Consumer protection rules: Consumer protection rules should not be affected, irrespective of who owns the credit and irrespective of the legal regime in force in the Member State of the credit purchaser or the credit servicer. Assigning a creditor's rights to a third party credit purchaser should not affect in any way the consumers’ right to plead against the assignee any defence, which was available to them against the original creditor.

·This means that an NPL platform would have to operate in full consistency with the Commission's proposal on credit servicers, credit purchasers and the recovery of collateral and therefore be compliant with Union law aimed at ensuring a high level of consumer protection, in the area of financial services. Furthermore, as regards the exercise of credit servicing activities in the cross-border context, the platform must operate in compliance with EU harmonised rules, which determine the applicable law, applicable jurisdiction and recognition, and enforcement of judgments in civil and commercial matters including in insolvency proceedings.

Finally, the proposal requirement that third-country purchasers of consumer loans use the services of EU authorised credit servicers and the additional safeguards in case of modification of the credit agreement would also have to be complied with.

·In line with the proposal on credit servicers, if adopted by the co-legislators, there would also be an accelerated extrajudicial collateral enforcement mechanism. According to the proposal, this would provide for a public auction or private sale framework. However, this relates to the realisation of the collateral attached to the non-performing loans in a portfolio (where those are secured by collateral). It does not relate to the sale of the actual loans. Therefore, this aspect of the proposed Directive would not affect the platform.

7.3.Data protection (data processing, data sharing, disclosure)

To the extent that the platform processes personal data it will have to fully comply with the GDPR. 39 Among other aspects, the following would need to be considered:

·a legal basis for the processing of personal data (e.g. law or contract);

·compliance with data protection principles notably purpose limitation, data minimisation, accuracy and storage limitation;

·availability of comprehensive information on data subject's rights including right to information, access, rectification, erasure, data portability, and with regard to automated decision making;

·clear definition of the controller or joint controllers (the banks selling the loans, the platform, both of them?) responsibility notably with regard to the principles of data protection by design and by default and the obligations on security of the processing, record of the processing activities, notification of personal data breaches and data protection impact assessment;

·definition of clear rules and justification on the legitimacy of sharing personal data with third parties;

·information on possibility to submit complaints to a data protection authority;

·compliance with rules on international transfers (transfers to third countries) of personal data.

The technical and organisational measures to be implemented by the platform to ensure compliance with the GDPR shall take be appropriate to the nature, scope, context and purposes of the processing as well as the risks it poses to the rights and freedoms of natural persons. In processing personal data the platform will also have to comply with the e-Privacy Directive (Directive 2002/58/EC, e.g. on cookie consent) and the forthcoming e-Privacy Regulation in replacing it.

In the context of NPL platforms, compliance with the GDPR can be supported by making use of safeguards such as encryption and pseudonymisation and the implementation of the data protection by default and by design principles.

7.4.Possible issues arising from insolvency and recovery regimes

The relevance of insolvency and pre-insolvency-related issues would depend on the functionalities that the platform would provide. In the work on defining industry standards for NPL platforms, it would be important to consider in detail the interaction between the platform’s operations and insolvency and pre-insolvency regimes in relation to the loans represented on the platform.

On a general note, and as stated elsewhere, the NPL platform would at no time own any loans, nor have any part in the settlement of transactions. The platform would primarily function as a data warehouse and help match potential sellers with buyers. The NPL platform would only have bilateral contractual links with the potential sellers, on one hand, and buyers, on the other. These contracts would regulate the provision of the platform’s services. Consequently, it would appear possible to organise the operations of the NPL platform in such a way that involvement of the platform in any potential insolvency and pre-insolvency procedures related to the NPLs represented on the platform could be avoided. However, in the event of the insolvency of a party any restrictions or obligations arising from the applicable law on insolvency and restructuring would have to be observed.

The platform could provide general information about insolvency systems and on the status of insolvency proceedings, to the extent practically possible. Since the enforceability of the loans present on the platform constitute an important feature for the pricing of these, the platform could make available up-to-date generic information about the loan enforcement and the relevant insolvency regimes. However, all parties would have to be made aware that the legal enforceability remains a characteristic of each individual loan, since besides statutory law, it may depend on agreements between the creditor and the debtor in the loan agreement.

7.5.Scope of legal documentation needed

The scope of the legal documentation needed will largely depend on the general scope of the NPL platform and primary and ancillary services offered.

At a minimum, the NPL platform would need to offer:

·Membership user agreements: each potential buyer and seller needs to sign a membership agreement to get access to the platform.

·Platform Rulebook: this includes rules on trading methods and terms and conditions for various actors, including sellers, buyers and third-party service providers, admission to trading, trading arrangements and the products (asset classes, single or bundled etc.). Before assuming their roles (putting assets for sale, or commencing bidding process), all users of the platforms would need to agree/sign the terms of conditions for their roles, including agreeing to the steps of the bidding process (receiving teasers, expressing interest, submitting a non-binding offer, submitting a binding offer etc.).

·Risk management documentation: a minimum level of risk management documentation, potentially via the membership agreement and platform rulebook, should include a standard non-disclosure agreement.

To the extent possible the above documentation, and in particular the non-disclosure agreement, should be standardised. However, it could be possible for the parties (buyers and sellers) to change the agreement if the need were to arise. The exchange of the signed documents (either electronically or physically) should be facilitated through the platform.

The platform should also specify legal documentation on individual loans to be provided by the seller to the buyers through the virtual data room.

Any deal-specific documentation, including actual sale-purchase agreements, would be outside the scope of the platform. Sellers would need to negotiate and sign a purchase-sale agreement with the successful buyer outside the platform, but within the timeframe specified by the platform rulebook. This is to simplify the process and requirements for the platform, and to ensure that the parties could express their specific agreement in the contract, including the choice of governing law.

8.Conclusions and potential next steps

In order to reap its full benefits, a European NPL platform should have a broad scope. All loan types should be eligible for inclusion on the platform. A European NPL platform should also be open to all types of sellers and professional buyers and have a wide geographical scope of operation, preferably EU-wide. In order to reduce transaction and search costs, the NPL platform should ensure data sharing and a high degree of data standardisation. To this end, an NPL platform should also, to the extent practically and legally feasible, help improve NPL data quality. Moreover, it should facilitate transactions by offering a price-setting mechanism of some sort for participants to use.

As described above, several private companies offer part of these services in a selection of Member States. However, they offer limited geographic scope and the loan data used is not standardised across the market. Finally, they are currently free to restrict access to sellers or for certain type of loans. For these reasons, among others, the potential benefits of a European NPL platform remain largely unrealised.

There is no clear-cut case for public ownership or operation of a European NPL platform for several reasons. Firstly, the recurrent operating costs of an NPL platform will likely dwarf the initial set-up costs. Therefore, in the case of public ownership, on-going public support covering platform operations could be needed. As there is no indication that a public operator would be more efficient than a private one, the rationale for public intervention in the running of the platform is weakened. Secondly, national ownership of a platform with a Union-wide scope may be problematic from a political as well as a fairness perspective. At Union level, it is unclear what the appropriate ownership vehicle would be in view of the authorities existing regulatory and executive roles. Therefore, and on balance, it appears preferable to further build on the existing private initiatives, rather than to overturn this embryonic market. Keeping this in mind, an operating model based on private law initiatives could offer many advantages, including avoiding budgetary implications.

European institutions would have to help industry experts to identify proper operating and governance standards for NPL platforms. These would be based on the principles outlined above and could take inspiration from existing Union law, where appropriate. The industry body would issue and enforce these requirements, and NPL platforms that meet these would be recognised as European NPL platforms.

In order to advance with the set-up of European NPL platform, a couple of steps could be taken without further delay. First, industry participants could be encouraged to start work on the industry standards and the setting-up of an industry body that would support the process of granting the ‘seal of approval’. Secondly, Union institutions as well as agencies and Member States could discuss whether further and stronger incentives may be needed to spur participation by sellers, especially banks, in an “approved” European NPL platform.

Annex 1: Examples of National NPL Platforms

NPL platforms with different functions exist across many Member States and have various shapes, functions, institutional setups and ownership structure.

In Portugal, a private coordination platform PNCB (Integrated Bank Credit Trading Platform) was launched in early 2018 by three major Portuguese lenders CGD, BCP Millennium and Novo Banco. The platform’s main objective is to manage the negotiation of non-performing claims and guarantees - that are common to at least two of the founding banks – on behalf of the lenders. PNCB is open to take other banks from the Portuguese banking system on board. The clear purpose of this platform is to maximise the value of the non-performing loans and possibly move the exposures back into the performing category or research the market for best bidders to allow for a swift sale of the credits. It stops well short of the transaction platform concept outlined later in this paper. The platform is intended initially to function for a period of three years. Nevertheless, this period is renewable, as the platform’s task is not expected to be finished within that short timeframe.

An online platform was also recently (back in summer 2017) setup by the well-established Spanish AMC SAREB. Its new strategy aims at both providing transparency to loan sales processes and opening them up to a much broader investor base than large institutional investors targeting big-ticket items. A key tool to implement this new strategy is the online platform. SAREB’s platform is targeting transactions and works in parallel with three of independent servicers, which also sell NPLs via their own online platforms.

In Greece, the high level of non-performing loans has led to the authorities and the banks to take a number of actions and initiatives that are being developed in parallel. In this context, the four systemic banks have established a common platform, called Project Solar, to focus on curing and maximizing recoveries from non-performing loans of SMEs that have exposures towards two or more banks. As per the functioning of this platform, the NPLs remain in the books of the banks whereas the banks and the servicer exchange information on a frequent basis to ensure its efficient functioning. Depending on the cooperativeness and the debt sustainability of each debtor, there are two available strategies for managing debtors, which are restructuring or liquidation.

The four systemic banks have also established a cross-bank coordination platform called ‘NPL Forum’ with the objective of achieving a closer cooperation in terms of loan restructurings of NPLs of corporates and SMEs with significant exposures towards at least 3 out of the 4 systemic banks. The NPL Forum meets regularly to discuss the cases of troubled borrowers, and to work on the agreement and implementation of common approaches.

(1) Fell, J., M. Grodzicki, D. Krusec, R. Martin, and E. O’Brien (2017), “Overcoming non-performing loan market failures with transaction platforms”, Special Feature A in Financial Stability Review, European Central Bank, November, and Box 7 in Financial Stability Review, European Central Bank, May 2018.

(2) Including in particular non-core and so-called IFRS 9 Stage 2 assets.

(3) COM(2017) 821.

(4) http://www.consilium.europa.eu/media/21594/leaders-agenda.pdf .

(5) The Commission issued a Communication [COM(2018)] on 7 March entitled ‘Completing the Capital Markets Union by 2019 — Time to accelerate delivery’.

(6) http://europa.eu/rapid/press-release_IP-16-3731_en.htm

(7) COM/2016/0723

(8) http://ec.europa.eu/finance/docs/policy/180314-staff-working-document-non-performing-loans_en.pdf

(9) COM/2018/0135

(10) COM(2018) 134.

(11) For example, some of those active first in the UK and Ireland and then in Spain, have entered the market in Italy or several Central and Eastern European Countries

(12) Apart from data of NPLs on banks’ balance sheets recorded by banking supervisors and central banks, which represent the potential supply of NPL, there are no official statistics to track NPL markets. Banks are not obliged to reveal them to statistical offices and often have no incentive to disclose details. Some consultancies collect data of individual sales from various sources and publish their information in reports. Information about realised market prices is generally not available, but treated as confidential by the parties involved in the transaction. Only a sample of transactions, mainly large, is disclosed.

(13) The number is measured in gross book value terms (unpaid primary balance). Since NPL prices are much lower than the gross book value of NPLs, both transaction values and amounts invested by non-banks are also smaller.

(14) Around 10 % in our sample.

(15) See Deloitte (2017).

(16) Prequin special report: Distressed debt in North America and Europe. 2016.

(17) In our sample 15 banks accounting for 12 % of the transaction volume.

(18) EBA Risk Dashboard Q2-2018.

(19) Fell, J., M. Grodzicki, D. Krusec, R. Martin, and E. O’Brien (2017), ‘Overcoming non-performing loan market failures with transaction platforms’, Special Feature A in Financial Stability Review, European Central Bank, November, and Box 7 in Financial Stability Review, European Central Bank, May 2018.

(20) SWIFT is a global member-owned cooperative and the world’s largest provider of financial messaging services.

(21) https://www.ifrs.org/about-us/who-we-are/

(22) https://www.ecb.europa.eu/paym/coll/loanlevel/html/index.en.html

(23) https://www.isda.org/

(24) https://www.globallei.com/

(25) In doing so, it would be essential to make sure that the European authorities’ involvement would be properly structured to avoid any legal implications of responsibility.

(26) Financial market infrastructures are allowed to establish links between themselves, and such links are widespread among central counterparties and securities’ depositories. See for example ESMA (2015), Final report on the extension of scope of interoperability arrangements, https://www.esma.europa.eu/sites/default/files/library/2015/11/2015-1067_-_report_on_io_extension_0.pdf .

(27) As defined in the EBA NPL templates

(28) This includes syndicated loans where banks and other lenders form a so-called syndicate to collectively meet the borrowing needs of a corporate client.

(29) Directive 2002/58/EC.

(30) However, the industry body could also consider promoting a hub-and-spoke model where the data warehousing function is centralised, but the remaining services are provided by various private entities, subject to licensing agreements that would ensure uniformity across all platforms. Akin to the concept of links between clearing houses and securities depositories, such a model would allow multiple transaction platforms to emerge and connect with each other, all operating on the same basis and accessing centralised data.

(31) It is noted that this idea would still need to be tested though the guidelines approval process, including public consultation, and, even if successful, would apply only to newly originated loans and will not affect the existing stock of NPLs.

(32) Legislative proposals on Amending Regulation (EU) No 575/2013 as regards minimum loss coverage for non-performing exposures and Directive on credit servicers, credit purchasers and the recovery of collateral.

(33) Directive 2014/65/EU and Regulation (EU) No 600/2014

(34) Regulation (EU) No 596/2014.

(35) 2011/61/EU.

(36) Directive 2013/36/EU, Regulation (EU) No 575/2013.

(37) Regulation (EU) 2017/2401 and amendment to Regulation (EU) No 575/2013.

(38) http://ec.europa.eu/finance/docs/policy/180314-proposal-directive-non-performing-loans_en.pdf

(39) Regulation (EU) 2016/679