Banks’ asset quality has substantially improved in recent years but remains well below European peers. Non-performing loans (NPLs) fell from 16½ percent in 2015 to about 8.1 percent at end-June 2019, achieved mainly through €145 billion of private NPL sales. This is a substantial reduction by any standard, though NPLs remain well above the 3.0 percent average of the main European Union (EU) banks as of June 2019. New NPL formation has fallen to pre-crisis levels. Provisioning coverage was 52.5 percent as of June 2019, placing Italy 7.6 percentage points above the average of the main EU banks.

Abstract

Banks’ asset quality has substantially improved in recent years but remains well below European peers. Non-performing loans (NPLs) fell from 16½ percent in 2015 to about 8.1 percent at end-June 2019, achieved mainly through €145 billion of private NPL sales. This is a substantial reduction by any standard, though NPLs remain well above the 3.0 percent average of the main European Union (EU) banks as of June 2019. New NPL formation has fallen to pre-crisis levels. Provisioning coverage was 52.5 percent as of June 2019, placing Italy 7.6 percentage points above the average of the main EU banks.

Executive Summary1

Banks’ asset quality has substantially improved in recent years but remains well below European peers. Non-performing loans (NPLs) fell from 16½ percent in 2015 to about 8.1 percent at end-June 2019, achieved mainly through €145 billion of private NPL sales. This is a substantial reduction by any standard, though NPLs remain well above the 3.0 percent average of the main European Union (EU) banks as of June 2019. New NPL formation has fallen to pre-crisis levels. Provisioning coverage was 52.5 percent as of June 2019, placing Italy 7.6 percentage points above the average of the main EU banks.

Intensive supervisory oversight in recent years has led to banks developing meaningful NPL management plans that include NPL reduction targets. The significant changes in the regulatory and supervisory landscape (see Box 3) has led to intense focus on banks’ NPL management strategies and reduction plans. Bank supervisors set up dedicated task forces to review and challenge the banks’ strategies for credibility and ambition as well as intensifying NPL reporting and monitoring of banks’ progress against targets. As a result, the significant institutions (SIs) with high NPLs are planning to reduce NPLs from €144 billion (or 8.3 percent of loans to customers) at end-June 2018 to €107 billion (or 7 percent of loans to customers) by end-2020, while the less-significant institutions (LSIs) aim to reduce the NPL stock from €21.4 billion (16.3 percent of loans to customers) at end-June 2018 to €13.3 billion (9.7 percent of loans to customers) by end 2021.2 Plans to reduce NPL ratios varies across banks, with the more profitable and better capitalized ones planning more aggressive reductions. Supervisors should continue with robust challenge of banks’ strategies and the ambition of their NPL reduction targets.

Banks’ plans to reduce NPLs are heavily reliant on disposals and write-offs due to limited possibility to obtain a rapid reduction through other means. On aggregate, banks’ projected volumes of NPL cures and internal workouts is almost matched by a new inflow of NPLs. This is reflective of the long delays with insolvency and enforcement procedures and of banks’ internal capacity constraints. It means that banks expect to reduce NPL levels mainly through sales and write-offs. The banks have achieved or exceeded the volumes of disposals planned up to December-2019, despite a general reduction in markets’ appetite for Italy country risk from spring 2018 to spring 2019.

Some complex features of the legacy NPL portfolio merits further supervisory investigation. Over 40 percent of banks’ current NPL gross stock is categorized as ‘unlikely to pay’ (UTP), reflecting banks’ continuing efforts to rehabilitate a large volume of distressed enterprises through restructuring.3 Successful rehabilitation may involve a mix of financial and operational restructuring, a change of business model and supply of fresh credit. Creditors face substantial challenges in this regard in case of small size firms, lack of reliable financial data, substantial multi-creditor issues, and the long time needed to reach a restructuring agreement. Considering the complex nature of these assets, bank supervisors are recommended to conduct targeted diagnostics using a representative sample of enterprises to ensure unviable firms are not being granted unsustainable forbearance.

Measures aimed at ensuring banks recognize losses on deeply delinquent debt using a calendar-based approach are welcome. Despite the substantial NPL reductions in recent years, the overall levels remain high and continue to weigh on bank profitability and thus banks’ market value and their ability to raise fresh private capital. The European Central Bank (ECB) and EU initiatives to gradually introduce calendar-based provisioning will ensure banks are incentivized to quickly restructure cases that can realistically be rehabilitated and recognize the costs associated with the recovery process for those borrowers that cannot.4 Calendar provisions will also help incentivize banks that have been so far slow in reducing their NPLs to do so more quickly. The Banca d’Italia (BdI) should consider extending the ECB’s approach that sets bank-specific expectations for the gradual path to full provisioning of the existing NPL stock to LSIs with high NPLs with an adequate phase-in period.

The BdI should consider more prescriptive guidance to LSIs on NPL management. The NPL guidance issued to LSIs in 2018 sets out expectations that banks should adopt formal policies for asset classification, forbearance, and valuation of assets. However, the LSIs have limited capacity to develop such policies and appropriately enforce high-level, qualitative regulation is always challenging. The guidance could usefully provide practical examples of when forbearance is appropriate; expect banks to use decision trees to guide forbearance decisions; exercise controls when applying multiple forbearance measures; expect banks to monitor and apply time constraints on any forbearance granted; provide examples of when loans should be valued using a going versus gone-concern valuation approach together with illustrative constraints on the collateral valuation methodology and minimum discounts and haircuts to be applied; and provide guidance on appropriate implementation of impairment triggers. In this regard, elements of the recently-issued European Banking Authority (EBA) guidelines on the management of nonperforming and forborne exposures could be used as a starting point to build-on.

Effective resolution of NPLs requires a well-functioning set of legal tools and institutional framework for debt collection, restructuring, and disposal. Recognizing the need to provide creditors with tools allowing speedy and optimal ways for credit recovery, Italy has undertaken several reforms of its legal system of debt enforcement and insolvency. The reforms aimed at allowing greater flexibility of the framework and enabling mechanisms for debt collection and restructuring. Informatization of courts (via telematics processes) have also brought notable improvements to judicial processes.

While recent reforms to the legal system have been positive further efforts continue to be needed. Realizing the benefits of the recent reforms requires considerable implementation efforts. These will include strengthening the regulation of the insolvency administrators, fine-tuning of the early warning mechanism, and improving the efficiency of liquidation processes. At the same time, further considerations should be given with regard to the revision of the current framework of creditor priorities and the reform of the special regime for large enterprises (e amministrazione straordinaria). Legislative changes are also warranted to provide greater legal certainty and flexibility for the out-of-court foreclosure mechanism.

Significant changes are still necessary in the institutional framework. The slow speed of judicial processes in Italy significantly contributes to the low recovery rates on NPLs and has a direct impact on the pricing of the NPL loans portfolios. The average time for the judicial enforcement of claims is five years while insolvency takes around seven years. There is a high heterogeneity in the duration of processes across the country. The long delays are due to lengthy in-court processes which are often due to the courts’ lack of resources and accumulated case backlogs. Enhancing the effectiveness of the judicial system would significantly benefit asset recovery in debt enforcement and insolvency cases and will have an important impact economy-wide. Improvements are necessary at the structural level of the Italian justice system in combination with further enhancements of internal court administration and case management. Judicial reform should focus on ensuring that courts have sufficient resources and expertise is spread evenly across the country.

Table 1.

Italy: Main Recommendations

article image
article image

C= continuous; I (immediate)= within one year; ST = Short Term (within 1–2 years); MT = Medium Term (within 3–5 years).

Tackling Problem Assets

A. The Italian NPL Landscape

1. Italian banks have made material progress in tackling the large volume of problem loans, but NPLs remain elevated relative to EU peers and continue to weigh on profitability. NPLs fell from €360 billion (16½ percent of gross customer loans) in 2015 to €190 billion (8.1 percent) in June 2019. This was achieved mainly through circa €145 billion of private NPL disposals. This is a substantial reduction by any standard, though NPLs remain well above the 3.0 percent average of the main EU banks as of June 2019. New NPL formation has fallen below pre-crisis levels. Provisioning coverage was 52.5 percent, placing Italy 7.6 percentage points above the average of the main EU banks as of June 2019.

Figure 1.
Figure 1.

Italian Banks’ Asset Quality Developments

Citation: IMF Staff Country Reports 2020, 234; 10.5089/9781513552156.002.A001

2. The largest share of gross NPLs (circa 77 percent) are concentrated toward small- and medium-enterprises (SMEs) and corporates. The aggregate NPLs by exposure are mostly concentrated in the services (31 percent by gross book value (GBV)) and construction (29 percent by GBV) sectors, with some sectors demonstrating significant vulnerability (e.g., 45 percent of credit to construction companies and a further 32 percent to real estate activities are non-performing). As of June 2018, approximately 52 percent of the total stock of gross NPLs was backed by collateral, with a further 20 percent having personal guarantees.

Figure 2.
Figure 2.

Italian Banks’ NPL Distribution

Citation: IMF Staff Country Reports 2020, 234; 10.5089/9781513552156.002.A001

3. More than half of the NPL stock is categorized as Sofferenze or bad loans, with the remainder being mostly unlikely to pay (UTP). The bank-client relationship with Sofferenze has been severed and efforts to recover value are exclusively through legal means. The remaining balance of NPLs, over 40 percent of banks’ gross NPL stock, is categorized mostly as UTP, reflecting banks’ continuing efforts to rehabilitate a large volume of distressed enterprises through restructuring (see Box 1).5 Successful rehabilitation often involves a mix of financial and operational restructuring, a change of business model, and supply of fresh credit. Creditors face substantial challenges due to the small size of the firms concerned, the lack of reliable financial data, the large number of creditors involved, and the long time needed to reach a restructuring agreement.

Unlikely to Pay (UTP) Portfolios

At €86 billion, the Italian banks’ UTP portfolios are substantial, accounting for over 40 percent of outstanding NPLs by exposure (42 percent for SIs and 48 percent for LSIs). The UTP portfolios are highly-concentrated with the four largest banks accounting for more than half of the total and 78 percent concentrated in the 10 largest Italian banks.

UTP loans are non-performing assets that have the potential to return to performing and thus the bank still maintains a relationship with the borrower. Loan loss provisioning coverage on UTP portfolios is 35 percent on average, compared to 69 percent for bad loans, reflecting the potential for the UTP borrower to rehabilitate. Lenders suffering from capital pressures are therefore incentivized to continue a relationship with failed borrowers while categorizing the exposure as UTP rather than as a bad loan.

The portfolios typically comprise of smaller loans (i.e., less than €10 million in size, with the majority being under €3 million) and the underlying borrowers typically are small enterprises that have loans with several creditors (up to 30 in some cases as reported anecdotally by industry practitioners). Lending to small and unsophisticated borrowers means that data quality and data continuity present challenges during a restructuring process, with banks often having to rely on qualitative rather than quantitative metrics to inform their restructuring decisions.

Market participants report that reaching agreement on a debt restructuring takes a long time (two to four years on average) due to the number of creditors involved and the variation in views on the way forward. The distressed enterprises concerned often receive insufficient fresh credit in the process, which undermines the rehabilitation effort.

The author of this box is Dermot Monaghan.

4. The Italian NPL market is the most active in Europe, with the 2018 volume of transactions exceeding €70 billion and estimated at €25 billion in 2019. The 2018 market volumes were driven mainly by very large deals (e.g., a €24.1 billion securitization completed by Banca MPS and a €10.8 billion portfolio sale by Intesa), securitizations backed by Garanzia Cartolarizzazione Sofferenze (GACS) (e.g., €5.1 billion by Banco BPM, €2.7 billion by UBI, €2 billion by BPER, €2 billion by ICCREA), and a number of true sale transactions. The 2019 volume of disposals is estimated to be about €25 billion, 30 percent of which was through the GACS securitization scheme (see Box 2).6

GACS: A Scheme to Support NPL Securitizations and Disposals GACS Overview

Garanzia Cartolarizzazione Sofferenze or “GACS” was introduced by Law n. 49/2016 and aims to protecting buyers of senior notes issued by a special purpose vehicle (SPV) by lowering their risk to that of a government bond, and to increase liquidity in the market by facilitating leverage on portfolio sales. The notes are backed by bad loans (known as “Sofferenze”) that are serviced by external servicers, which are independent from the originating bank. The intervention of the Italian Government is limited to the coverage of the interest and capital payment obligation on the senior tranches of notes. The guarantee can be called on the amount outstanding on the senior notes at their legal final maturity date.

Main Features

A set of criteria governs the eligibility for the senior notes to receive the guarantee. Among these: The senior notes need to be rated investment grade and the rating is not revocable; The bank has to sell at least 51 percent of the junior tranche; It must obtain deconsolidation and de-recognition of the assets sold; The bank cannot be affiliated with the servicer; The waterfall allows for the repayment of the principal on the senior to be subordinated only to the payment of interest on the mezzanine (if issued), but not the junior notes; and the premium for the guarantee is a senior cost in the waterfall and, to assure the aid-free nature of the scheme, is at a market-rate: based on the average price of a basket of single name CDS covering investment grade rated Italian companies, for the same duration of the notes.

Market impact

Banks had completed 21 transactions using the GACS scheme by March 2019, securitizing more than €62 billion in bad loans. There was just one GACS-backed transaction completed in 2016 (Banca Popolare di Bari, €480 million GBV), reflecting the long time needed for data tape analysis and for the senior tranche to receive a credit rating. The scheme was used more widely for non-performing exposure (NPE) securitizations in 2017 (several transactions, over €7 billion GBV) and 2018 (14 transactions worth circa €45 billion GBV). The scheme has been extended three times: in September 2017 for one year, in September 2018 for six months, and in March 2019 for two years.

Extension to UTP

Market commentators have argued that the scheme should be extended to UTP portfolios, however there are several technical challenges including: ratings agencies have yet to develop the models to rate tranches of UTP exposure; valuation of highly granular UTP portfolios is expensive; difficulties with standardizing the data tape for investor due diligence; the operating relationship with debtors needs to be maintained; and multi-creditor challenges including achieving a critical mass of creditors in the securitization to effect enterprise rehabilitation.

The author of this box is Dermot Monaghan.

5. Some banks may not have sufficient loan loss provisions to decisively tackle NPLs and meet NPL reduction targets. A high-level review of banks’ loss experience (see Box 3) identifies variations in provisioning coverage across banks that are not fully explained by portfolio characteristics and many banks’ provisioning coverage is lower than the observed loss experience. The analysis identified a range of €5 billion (if NPLs are worked out internally) to €20 billion (if NPLs are disposed) of further provisioning for the SIs and 53 LSIs considered in the analysis. Most of the additional provisioning needs is attributed to the UTP portfolios (see Table 2). Considering banks have NPL reduction plans and NPL reduction will only be achieved through disposal, loan loss provisions would likely have to increase by about €7.2 billion in order to achieve the 2018 agreed reduction targets.7

Table 2.

Italy: Potential Additional Loan Loss Provisions Needed for SIs and LSIs

article image
Source: IMF staff estimates.

Adequacy of Provisions to Support NPL Reduction

Data Review

The mission collated information from several sources, including EBA transparency exercises and risk dashboards (which apply to the SIs only), SI’s financial and other public statements, meetings with individual banks, ECB Finrep data (for SIs only), Bank of Italy FSAP data for LSIs (using a representative sample of 62 banks), Published BdI aggregate data (including bad loan recovery statistics, financial stability reports, monetary & financial statistics, and occasional notes), Recovery information from NPL servicers and market advisory/consultants proprietary data.

Dominant Drivers of Credit Loss

Citation: IMF Staff Country Reports 2020, 234; 10.5089/9781513552156.002.A001

Key Observations

The data identified two dominant drivers of loss – the use of credit risk mitigation and the legal status of the NPL. The data also identified that over recent years banks have demonstrated a consistent finite capacity to resolve NPLs through internal means and thus NPL reduction will realistically only take place though disposals or write-offs–an observation consistent with banks’ NPL reduction plans. The gap between banks’ net book values and market values remains high at 13 percent on average, which has been attributed to the high uncertainty in the value recovery process (most of the recovery in Italy is through creditors settling with debtor at a substantial discount to the loan balance) and distressed debt investors command a substantially higher return on capital than banks.

Implications for Loan Losses and Adequacy of Provisions

On average, banks have recovered 38 percent on bad loans through internal workout and 58 percent for UTP and past-due portfolios. The value recovery falls to 25 percent for bad loans and 43 percent for UTP and past-due once banks dispose of the risk in the open market. However, many banks loan loss provisions fall below these averages, in particular for UTP portfolios.

Losses from Bad Loans

Citation: IMF Staff Country Reports 2020, 234; 10.5089/9781513552156.002.A001

Losses from UTPs & Past Due

Citation: IMF Staff Country Reports 2020, 234; 10.5089/9781513552156.002.A001

The author of this box is Dermot Monaghan.

B. Recent Actions and Reforms

6. Intensive supervisory oversight in recent years has led to banks developing meaningful NPL management plans that include NPL reduction targets. The significant changes in the regulatory and supervisory landscape (see Box 3) has led to intense focus on banks’ NPL management strategies and reduction plans. Bank supervisors set up dedicated task forces to review and challenge the banks’ strategies for credibility and ambition as well as intensifying reporting for NPL portfolios and monitoring of banks’ progress against targets (see Box 4). As a result, the SIs developed NPL reduction plans that they have successfully met to date. The SIs are planning to reduce NPLs from €144 billion at end June 2018 to €107 billion (or 7 percent on average) by end-2020, while the LSIs aim to reduce the NPL stock from €21.4 billion (16.3 percent) at end-June 2018 to €13.3 billion (9.7 percent) by end 2021 (see subsequent paragraphs for more details).8

7. The BdI followed the ECB’s lead and issued guidance on the management of NPLs for LSIs in January 2018. The guidance calls on banks to adopt a formal strategy for optimizing NPL management by maximizing the current value of recoveries. It asks banks to develop short and medium to long-term operational plans (of roughly one and 3/5 years respectively) for managing NPLs, setting out operational targets for NPL reduction. The guidance also included expectations on the governance and operational arrangements for NPL management as well as expectations for banks to develop policies on forbearance measures; loan classification; value adjustments and write-offs; and valuation of real estate collateral.

Regulatory and Supervisory Initiatives to Improve Asset Quality

  • The 2014 EBA definition of Non-performing Exposures (NPEs) and forbearance and the 2018 EBA guidelines on disclosure of non-performing and forborne exposures;

  • The 2014 ECB Comprehensive Assessment, a combination of asset quality review and solvency stress test applied to the largest Italian banks that became under direct supervision by the then newly-established Single Supervisory Mechanism (SSM);

  • The Capital Requirements Directive 4 package (CRD4/CRR) strengthened the prudential regulation by implementing the main elements of Basel III in the EU (2014);

  • The adoption of EU 2016/867, the AnaCredit Regulation on the collection of granular credit and credit risk data and the related amending decision on the organization of preparatory measures for the collection of granular credit data by the European System of Central Banks;

  • The adoption of EU 2016/2067, the application of IFRS 9 from January 2018, which requires the measurement of impairment loss allowances to be based on an expected credit loss accounting model rather than on an incurred loss accounting model; The impact on banks’ capital ratios is being smoothed in line with EU’s transitional provisions.

  • SSM guidance to banks on the management of non-performing exposures provides qualitative guidance and specifies supervisory expectations for the provisioning of NPLs (2017/2018/2019);

  • The EU Council’s 2017 Action Plan includes several new measures including a prudential backstop to ensure minimum levels of loan loss provisioning; EBA guidance (2018) on NPL management; guidelines on banks’ loan origination, monitoring and internal governance practices; macro-prudential approaches to prevent the emergence of system-wide NPL problems; enhanced disclosure requirements on asset quality and non-performing loans; guidelines for banks on loan tapes monitoring; standardized data for NPLs and NPL transaction platforms; and an approach to foster the development of secondary markets for NPLs; and

  • The Guidance on the management of non-performing exposures for Italy’s LSI published by BdI explains supervisory expectations for the management of NPLs by LSIs (2018).

The author of this box is Dermot Monaghan.

8. Following the submission of NPL strategies from high NPL LSIs in September 2018, the BdI has reviewed these strategies and provided its feedback to the relevant LSIs. The BdI has requested 53 LSIs with high NPL ratios (other than cooperative banks, also known as Banche di Credito Cooperativo (BCCs)) to define and prepare or update their NPL management plans by December 2021.9 The BdI established a dedicated task force to assess the plans for credibility and ambition. Based on these plans, the NPL ratio for a sample of 47 LSIs is expected to decrease from 16.3 percent in June 2018 to 9.7 percent in December 2021. The LSI’s planned reduction of NPL stocks by end 2021 is similar to SI’s targets by end 2020. The BdI considers that the lower targets defined in the NPL strategies by LSIs are relevant given the greater difficulties faced by LSIs in internal workouts and NPL sales. The credibility of the plans and the consistency of some assumptions need to be monitored and verified given the current macroeconomic context.

9. The process of challenging and monitoring the progress of the LSIs’ reduction plan is ongoing. As part of the assessment of the NPL reduction plans, the BdI identified some banks where the plans were not sufficiently ambitious or reliable. As a result, the BdI requested amendments to these plans. The second round of the plan updates was scheduled for March 2019, and an annual update will take place afterwards.

10. While the NPL guidelines are a welcomed initiative, they are not binding and are not detailed enough particularly in relation to forbearance measures and processes. The BdI’s concise guidelines invites the LSIs to refer to the SSM Guidance for the operational details. Incorporating prescriptive expectations in the guidance would ensure more consistent approach by banks and supervision teams, especially relating to forbearance, collateral valuation and impairment triggers. Including indications for the viability and the timeline of forbearance measures and the main elements of a sound forbearance process, as outlined in the recent EBA guidelines on the management of non-performing and forborne exposure issued in October 2018 would also help.

11. While banks’ plans to reduce NPLs are heavily reliant on disposals and write-offs due to limited possibility to obtain a rapid reduction through other means. On aggregate, both SIs’ and LSIs’ projected volumes of NPL cures and internal workouts is almost matched by a new inflow of NPLs. This is reflective of the long delays with insolvency and enforcement procedures and of banks’ internal capacity constraints. It means that banks expect to reduce NPL levels mainly through sales and write-offs. The volumes of disposals currently planned appear achievable in the current liquid market environment but any reduced foreign investor appetite for Italy country risk could have an adverse impact. Plans to reduce NPL ratios varies across banks, with the more profitable and better capitalized ones planning more aggressive reductions.

Figure 3.
Figure 3.

Banks’ NPL Reduction Plans

Citation: IMF Staff Country Reports 2020, 234; 10.5089/9781513552156.002.A001

Source: ECB/SSM and Banca d’Italia.

C. Policies to Promote NPL Reduction

12. Supervisors should continue to prioritize strategies and reduction plans for banks with high NPLs. Recent experience confirms that persistent and assertive supervisory oversight of banks’ NPL management strategies and reduction targets results in bank owners and management taking credible steps to materially reduce the adverse impact of NPLs on the banks’ long-term viability. This assertive supervisory approach is welcome and needs to continue for all Italian banks suffering with high NPL levels.10

13. The BdI should further enhance its regulations and guidelines to LSIs relating to the treatment and management of problem assets. LSIs have limited capacity to develop complex NPL management policies and enforcing high-level, qualitative regulation is always challenging. In this regard, a more prescriptive approach can be helpful, and the following recommendations are proposed:

  • The BdI should review its NPL guidelines (considering also transforming it into a binding regulation) to incorporate the EBA NPE definition and subsequent guidelines on the management of NPEs (see Box 4), with a view to prescribing more detailed criteria and examples of: when forbearance is appropriate; expect banks to use decision trees to guide forbearance decisions; exercise controls when applying multiple forbearance measures; expect banks to monitor and apply time constraints on any forbearance granted; provide examples of when loans should be valued using a going versus gone-concern valuation approach together with illustrative constraints on the collateral valuation methodology and minimum discounts and haircuts to be applied; and provide guidance on appropriate implementation of impairment triggers. The BdI should continue monitoring the implementation of banks’ NPL plans and scrutinizing banks’ NPL plans for credibility and ambition.

14. Considering the complex nature of Italian UTP exposures, supervisors are recommended to pay particular attention to banks’ management of the risks therein. The scale and features of Italian banks’ UTP portfolios (see Box 3) merits caution to ensure that credit risks are fully identified and adequately provisioned for. The key challenge is to confirm if the enterprises being considered by banks for financial restructuring are credibly demonstrating the potential to become viable in the long term. In that regard, bank supervisors are recommended to:

  • Undertake targeted on-site reviews to ensure that unviable firms are not being granted unsustainable forbearance. Given that assessing firm’s viability involves industry and sector-specific expertise together with detailed data, it is recommended that banking supervisors consider partnering with one or more specialist firms for a one-off, limited-term project aimed at identifying existing balance sheet vulnerabilities and building the methodological approach for future on-site challenges.

  • Develop guidance for banks and review internal supervisory policies to steer on-site inspections. The expectation should be that credit exposure categorized as UTP should fulfill some minimum criteria to materially separate it from a bad loan and avoid attracting provisions more consistent with same. Examples include: the enterprise is operating as a credible going concern entity (i.e., demonstrates clear evidence of generating cashflows sufficient to service debt under a lower interest rate and longer time horizon, and at a minimum, having an interest service coverage ratio of greater than 1); the firm has sufficient current and historical quantitative financial data available to credibly inform a financial restructuring; the firm is fully cooperating with all creditors’ requests and is forthcoming with pertinent information and sharing with all other creditors; no creditor has litigated against the firm in in a way that threatens a potential financial restructuring; the firm is not subject to third-party (non-creditor) litigation that places the viability and solvency of the firm at material risk; the loan has been categorized as UTP for less than 18 months; the loan has been categorized as UTP for more than 18 months but with clear evidence that more than half the creditors by total credit exposure are supportive of a proposed restructuring solution and are in cooperation with each other; and for medium and large firms (>50 employees, turnover >€10 million, and balance sheet >€10 million), the enterprise has been subject to an independent business review to verify viability and is undergoing a mix of financial and operational restructuring with the involvement of industry experts.

15. Measures aimed at ensuring banks recognize losses on deeply delinquent debt using a calendar-based approach are welcome and needed. Despite the substantial NPL reductions in recent years, the overall levels remain high and continue to weigh on bank profitability and thus banks’ market value and their ability to raise fresh private capital. The ECB and EU initiatives to gradually introduce calendar-based provisioning will ensure banks are incentivized to quickly restructure cases that can realistically be rehabilitated and recognize the costs associated with the recovery process for those borrowers that cannot.11 Calendar provisions will also incentivize banks that so far have been slow in reducing their NPLs to do so more quickly. The BdI should consider extending the SSM’s approach that sets expectations for the gradual path to full provisioning of existing NPL stocks to LSIs with high NPLs with an adequate phase-in period.

Figure 4.
Figure 4.

NPL Ratio vs. Market Cap/TBV for Exchange Listed Italian Banks

(Bubble size: Banks’ total assets)

Citation: IMF Staff Country Reports 2020, 234; 10.5089/9781513552156.002.A001

Sources: SNL, Reuters

Legal Aspects of Npl Resolution

16. This part of the paper discusses the selected legal aspects of insolvency and creditor rights that are immediately relevant to resolution of distressed corporate debt. The focus on the corporate insolvency and creditor rights framework is motivated by the concentration of the largest part of the NPL portfolio in Italy in corporate sector.12 The discussion is not a full assessment of the Italian insolvency and creditor rights regime against the international standard.

17. Effective resolution of NPLs requires a well-functioning set of legal tools for debt collection, restructuring, and disposal. There are five key approaches to tackle NPLs, from a legal and economic perspective (see Figure 5), i.e., enforcement of collateral, insolvency, loan restructuring, settlement with debtor at a discount to the loan value, and disposal of risk to third parties. The foundation of NPL resolution is based on the two key pillars of enforcement and insolvency. Efficient insolvency and enforcement mechanisms also push creditors to negotiate restructuring solutions as these solutions are negotiated “in the shadow of the law” (i.e., debtors and creditors negotiate knowing what the outcome would be in the case of resorting to a legal process). Finally, the prices for the sale of NPLs are heavily influenced by the duration, cost, and effectiveness of the debt recovery and restructuring mechanisms.

Figure 5.
Figure 5.

Banks’ NPL Resolution Options

Citation: IMF Staff Country Reports 2020, 234; 10.5089/9781513552156.002.A001

Source: IMF staff.

18. Italy has introduced a number of reforms to improve its debt enforcement and insolvency frameworks. Recognizing the need to provide creditors with tools allowing speedy and optimal ways of credit recovery, Italy has undertaken a number of reforms to its legal system of debt enforcement and insolvency. In particular, the 2015–2016 amendments to the enforcement framework focused on shortening procedural deadlines and enhancing the flexibility for price formation in auctions and transparency in sales. A number of changes to the insolvency system over the period 2006–201813 significantly modernized the framework, culminating in the new insolvency law adopted in January 2019 and expected to take full effect in August 2020.

19. Despite recent reforms, the slow speed of judicial processes in Italy continues to be a key reason for low NPLs recovery rates and the large discounts at which assets are sold.14 According to the data of the BdI,15 the average recovery rate for total NPL positions closed in 2017 (including both internal workout methods and sales) stood at 30 percent, with the average recovery rate for all NPLs secured by collateral at 39 percent, and unsecured credit at 21 percent. At the same time, the average time for real estate judicial enforcement of claims based on 2017 data is five years, while insolvency processes on average take somewhat over seven years. The long delays in recovery and the related uncertainty have a direct impact on the banks’ internal NPL management as well as on the pricing of NPLs’ portfolios.

A. Judicial Enforcement

20. Delays of judicial processes in Italy are endemic, but also highly variable from court to court. Collection of secured credit through the foreclosure procedure (procedura esecutiva immobiliare) takes on average five years (see Table 3). The general inefficiencies of the judicial system also affect the enforcement of unsecured debts. The length of proceedings is highly heterogeneous among different courts and regions (see Figure 6). The high court-to-court variance of the length of the enforcement processes can be attributed to the different levels of courts’ efficiency, insufficient resources (especially, human resources),16 and the existing backlogs of the courts. Foreclosure is performed in several stages and starts with a court order for seizure of assets, followed by appraisal and the auction sale conducted by the public notary (or other professional) under court supervision, concluding with the distribution of proceeds to the creditors. For unsecured credit, the use of the ordinary procedure results in even lengthier processes as it includes verification of the legitimacy of the claim, followed by the identification of the debtor’s assets for seizure and sale. The announced Civil Procedure reform is yet in the works, and, as previous reforms, its objective is to streamline and simplify the judicial processes at all levels.

Table 3.

Italy: Real Estate Foreclosures: Length and Number of Closed Files

article image
Note: 2018 numbers are based on data as of July 31, 2018.Source: Cerved (2019).
Figure 6.
Figure 6.

Length of Foreclosures by Court

Citation: IMF Staff Country Reports 2020, 234; 10.5089/9781513552156.002.A001

Source: Cerved (2019).

21. Some evidence suggests a recent reduction in the duration of foreclosure procedures. The 2015–2016 reforms, inter alia, reduced the procedural deadline of the pre-sale stage of foreclosure, improved procedural rules on the conduct of auction sales, and allowed more flexibility in auction price formation by allowing reductions of up to 50 percent from the estimated value of assets.17 These reforms brought some shortening in the duration of foreclosure processes with the biggest impact measured at the auction phase.18 At the same time, the post-sale stage of the foreclosure (distribution of proceeds to creditors) has not been affected by the reforms, and market participants continue to report long delays between the sales and the distribution of proceeds, which could take many months. These delays in the distribution of proceeds are often attributed to the courts’ internal processes in managing the distribution phase.

22. The introduction of the nation-wide online platform for sales of assets could further improve the collateral market. The 2016 changes to the Civil Procedure Code provided that forced sales of assets must be advertised in the single national on-line platform managed by the Ministry of Justice.19 The new system became effective in February 2018. The objective of the platform is to increase transparency and participation in public sales of assets. The platform also serves as the interface between the potential buyers and the private auction operators which are appointed at the court level to manage the auction process.

23. The online platform is at its early stages of development. Market participants report that the system does not yet represent the intended “national market place” for advertisement of the assets. The information on the assets may be hard to locate as it is not standardized in a user-friendly way.20 Creditors continue to use alternative means of advertisement with private agents at the local level. The platform is not designed as an auction platform but rather provides links allowing registration with the local auction operators who are registered for the conduct of such activity with the Ministry of Justice. Auction operators are chosen via the roster system at the individual court level. The system is not harmonized, and different courts operate rosters under their internal rules.

24. The time of foreclosure continues to vary significantly across the country. On average, Northern Italy registers a higher level of court efficiency and reduced case backlogs, compared to the South, Center, and Islands (see Table 4). There are a number of structural factors which explain the differences in performance,21 including the current system of appointment and rotation of judges which results in a higher concentration of more experienced judges in the Northern regions. Furthermore, backlogs at the level of courts of appeal and the Supreme Court are reportedly significantly larger compared to those at the ordinary courts. To help the process of harmonization and improvement of court practices, the National Justice Counsel introduced the 2017 guidelines on best practices for the foreclosure process with an aim to harmonize and encourage faster foreclosure processes (e.g., including by providing case management and organizational recommendations with regard to the conduct of assets sales and encouraging the partial distribution of auction proceeds when possible). The effect of the guidelines is yet to be seen.

Table 4.

Italy: Average Length of Foreclosures by Geographical Area

article image
Source: Cerved (2019).

Recommendations

  • Improvements are necessary at the structural level of the Italian justice system in combination with strengthening resources, organization, and case management at the court level. Judicial reform should focus on improving the structure of the judiciary and ensuring that expertise/experience is spread evenly across the country. The specialization of judges on commercial matters should be strengthened. Civil procedure rules require streamlining and simplification for greater efficiency of processes.

  • Additional changes to the foreclosure procedures could be considered to reduce the time between asset sale and the distribution of proceeds to creditors. Further standardization in court management of practices in the distribution of proceeds is required. Consideration should be given to transferring the duties of managing the stage of distribution of proceeds from the judge to the court representative subject to judicial control.

  • The online public platform for auctions needs further development. Improvements to the IT system, including the standardization of asset-related data in a user-friendly manner, are necessary to ensure that the platform serves as nation-wide market place for public assets sales. This needs to be combined with harmonization of the court practices on the choice of local auction operators to bring uniformity and predictability.

B. Out-of-Court Enforcement of Real Estate Collateral

25. The out-of-court enforcement mechanism for real estate collateral, introduced in 2016, has not taken off. The May 2016 law gave financial creditors of defaulted corporate debtors the right to directly repossess the collateral in lieu of debt payment if a respective clause (so-called “patto Marciano”)22 is foreseen in the loan agreement. The intention of the mechanism is to speed up enforcement by avoiding the delay-ridden judicial enforcement system. The ability to use the out-of-court enforcement depends on the inclusion of the respective clause in the loan agreement, so the potential benefits of the mechanism apply to future NPLs rather than existing stocks, as it is unlikely that a borrower of a defaulted loan would agree to the inclusion of the clause in the loan agreement without extracting concessions from the lender. However, banks reported to continue to be reluctant to introduce the Marciano clause in their loan contracts and the practice has not spread among the industry. This is despite the October 2018 changes to the prudential regulation which relaxed the limits on the banks’ holdings of repossessed assets.

26. The lack of legal certainty, rigid repossession rules, and reputational risks have impeded the banks from taking advantage of the Marciano mechanism. The banks are reluctant to repossess collateral both due to reputational concerns and to the additional burden that such repossession would entail related to asset management. While repossession is based on the appraisal value and any disputes over it do not block the repossession process, the current law regulating the Marciano mechanism lacks clarity on the rules applicable to disputes over the appraisal value, which significantly limits the mechanism subjecting the creditor to the risk of inaccurate appraisals. Furthermore, the law does not envisage the possibility to opt-out from the repossession procedure if the evaluation of the real estate provided by the valuator is deemed unsatisfactory. To promote the mechanism, the Italian Banking Association and Confindustria signed a memorandum of understanding (MOU) aimed at addressing the perceived reputational risks and providing for a common understanding of some of the provisions of the law.23 However, as the memorandum does not have the force of law, it is not sufficient to overcome the perceived shortcomings of the out-of-court enforcement mechanism and, more importantly, it cannot go far enough to ensure the necessary flexibility of its use.

Recommendation

  • Further flexibility and clarity in the regulation of the out-of-court enforcement mechanism is needed to incentivize its use within the constraints of constitutional law. To facilitate out-of-court foreclosure the law could explicitly allow creditors the option to sell the collateral, in addition to the repossession. This would require further changes to the current framework, including the regulation of the sale process, and perhaps allowing some margin for price reduction below the appraised value, in a public sale.24 For the use of the repossession option the law should introduce detailed rules for the appraisal disputes and further address the rigidities and uncertainties identified in the MOU between the Italian Banking Association (ABI) and Confindustria.

C. Debt Restructuring

27. Multi-creditors negotiations in Italy reportedly take many months and there are no guidelines to assist this process. A set of principles and guidelines that govern the negotiation and conduct of participants in multi-creditors negotiation, like those adopted by INSOL International (Global Approach to Multi-creditors Workouts) could be useful even in the absence of their binding effect. While the ABI considered this matter in the past, the project to adopt debt restructuring guidelines never materialized.

28. The current system provides for several pre-insolvency out-of-court restructuring tools which are widely used. The pre-insolvency restructuring mechanisms are covered under the insolvency framework law and include out-of-court work-out plans (“piani attestati di risanamento”) and restructuring agreements (“accordi di ristrutturazione”), with a special type of restructuring plans designed for financial creditors.

29. The workout plans (piani di risanamento) are “private” agreements fully outside of court control. These plans are not subject to obligation to be disclosed to public sources and can provide for an internal reorganization (e.g., merger or capital increase) or involve restructuring agreements with creditors. These are certified by an independent expert appointed by the debtor and have the advantage to provide protection for the acts executing the plan from the regime of avoidance actions and criminal prosecution in case of subsequent bankruptcy proceeding. The 2019 insolvency law strengthened documentary requirements for the piani di risanamento to ensure more safeguards against potential abuse.

30. The restructuring agreements (accordi di ristrutturazione) are subject to court control. Such agreements can be concluded when agreed with creditors representing at least 60 percent of all claims against the debtors. Special rules for restructuring agreements with financial creditors allow the extension of the effect of the restructuring agreement on dissenting creditors when approved by 75 percent of all financial creditors’ claims. The advantages for the restructuring agreement include the recognition of the priority of new finance accorded within the scope of the agreement and the possibility to request the stay on creditors’ actions during the negotiation of such agreements. The court’s control and validation of restructuring agreements is principally aimed to verify that the interests of non-adhering creditors are not impaired. Following filing of such agreements with the court, the debtor publishes the restructuring agreement in the companies’ register where the debtor company has its registered office.

31. The 2019 insolvency reform introduces changes on the regulation of debt restructuring agreements. The 2019 law (discussed in greater detail below) integrates the institute of the out-of-court restructuring agreements into the broader framework of insolvency, providing that in case of failure to approve such agreements the case could enter into the in-court concordato or liquidation stage. It also relaxes, in certain cases, the requirements for the conclusion of restructuring agreements. The new law lowers the approval threshold for such agreements from 60 percent to only 30 percent of creditors’ consent provided interests of non-adhering creditors are not affected. This provision however applies only in exceptional circumstances, i.e., when the agreement does not propose a stay on creditors’ actions, if a stay is proposed the 60 percent threshold continues to apply. The regime of the current “financial restructuring agreements” with its cramdown mechanism, which applies only to financial creditors, will be extended to other classes of creditors. This would allow agreements approved by 75 percent of the creditors’ claims of a class to be binding on the dissenting creditors of the same class.

32. Among the major novelties of the new law is the introduction of an early warning system. The objective is to incentivize early restructuring at a time when insolvency could still be avoided, thus addressing the long-standing issue of late insolvency filings. The new procedures require companies to take necessary actions to address a situation of distress that will be assessed as per the set of financial distress indicators which are yet to be adopted. More companies are now required to create internal control bodies, which increases the costs of doing business for small enterprises. The law also puts an obligation on debtor’s statutory and external auditors and certain categories of public creditors (e.g., tax administration)25 to immediately inform the company’s directors of the signs of a crisis, and in case of failure to address it, they are required to turn to a special “crisis-assistance commission” (so called Organismi di Composizione della Crisi (OCRI)) to be created at the level of the local Chamber of Commerce26 which are meant to assist the debtor in working out a consensual arrangement with its creditors. In case that an agreement with creditors is not reached within 6 months (as of the date of the start of the OCRI-led process) the OCRI reports the case to the public prosecutor which can file the insolvency (liquidation) against the debtor ex officio. The use of the early warning mechanism is incentivized by an exemption or decrease of liability for certain insolvency crimes, the reduction of tax sanctions, and a delay of the duty to request the commencement of insolvency. These new procedures are meant principally to assist small and medium size companies and do not apply to large companies or listed companies.

33. Meeting the policy objectives of the early warning system may be challenging and will heavily depend on future implementation. Italy has no experience with the early warning system introduced by the new law and there are no best practices for the early warning mechanism. The Italian version of the mechanism appears rather complex and highly administrative. The potential involvement of the prosecutor to file for the debtor’s insolvency needs to be carefully considered. The efficiency of the process seems to rely heavily on the ability and competence of the OCRI (yet to be created) to assist in the context of corporate distress. The appropriateness and flexibility in the use of “crisis indicators” (yet to be adopted) triggering the early warning mechanism will be crucial. Potential risks include the perceived reputational risks of negative signaling to creditors and business partners of the affected enterprise. Another risk could be that of creating an antechamber to insolvency which delays insolvency filings, with potential avoidance of liability.

Recommendations

  • Promote the development of a code of conduct on a voluntary and consensual for debt restructuring. This could be done through the ABI with further endorsement and assistance of the BdI.

  • The new mechanism of early warnings could benefit from further refinements to ensure its efficiency. Further considerations are warranted to avoid negative signaling which may be associated with entrance into early warning processes. The law and implementing regulations should ensure that the passage through the mechanism does not replace the timely submission of an insolvency filing. In this regard the crisis indicators which are yet to be adopted and which will be used to trigger the early warning procedures should be sufficiently flexible to accommodate the variety of corporate distress situations and should allow room for judgement as to whether the use of the early warning mechanism is useful. While experience is gained, the strict mandatory nature of this mechanism could be reconsidered in favor of a more voluntary type of procedure. Consideration should also be given to simplifying OCRI processes, issuing guidelines for harmonization of OCRI-assistance and building up the competence of OCRIs.

D. Enterprise Insolvency

34. The efficiency of the current insolvency system is undermined by complex and long procedures. The regime is based on the 1942 insolvency law which has been subject to multiple changes over the last decade27 in an attempt to modernize the framework. Such piecemeal changes have added to the complexity of the system compromising its transparency and efficiency. The average length of the insolvency procedures according to the latest statistics is around seven years (see Figures 7 and 8 and Table 5) with a high variability of length from court to court (see Figure 9). The insolvency filings often come at a stage where saving the business is impossible. The restructuring mechanisms are poorly integrated into the general insolvency framework. The liquidation process is slow. The system is also marked by the existence of numerous priorities in favor of certain creditor groups—including public creditors. The court delays and the lack of judicial specialization in restructuring and insolvency further aggravate the inefficiencies of the system.

Figure 7.
Figure 7.

Average Length of Insolvency Procedures

Citation: IMF Staff Country Reports 2020, 234; 10.5089/9781513552156.002.A001

Source: Cerved (2019).
Figure 8.
Figure 8.

Insolvency Case Status History

Citation: IMF Staff Country Reports 2020, 234; 10.5089/9781513552156.002.A001

Source: Cerved (2019).
Table 5.

Italy: The Average Length of Pending Insolvency Cases

article image
Source: Cerved (2019).
Figure 9.
Figure 9.

Average Length of Closed Insolvencies by Court–2018

Citation: IMF Staff Country Reports 2020, 234; 10.5089/9781513552156.002.A001

Source: Cerved (2019).

35. The latest insolvency reform focuses on the rationalization of the insolvency system in an attempt of realigning it further with international best practices. The 2019 reform providing for a single procedure for insolvency cases based on broad commencement criteria: a state of “crisis” which includes both actual and imminent insolvency. A single procedural scheme will apply independently of the eventual choice of restructuring or liquidation options.28 Informatization of certain processes, including the introduction of unifying different processes by electronic voting, helps to speed up some of the insolvency procedural actions. This is expected to reduce the procedural delays related to transitioning from one stage to another. In an effort to shorten processes and in line with best practices, the duration of the insolvency moratorium has been limited to 12 months.

36. The proclaimed intention of the reform is to achieve a better alignment with EU and international insolvency standards. The new law adopted in January 2019 is based on the proposals of the Rordorf Commission, created by the government and composed of insolvency experts. The design of the new law was guided by the requirements of EU Regulation 2015/848 on Cross-border Insolvency and the principles of the 2014/135/EU: Commission Recommendation on a New Approach to Business Failure and Insolvency, in addition to the insolvency recommendations from UNCITRAL. The main theme of the revisions is the encouragement of early restructurings, the avoidance of insolvency, and the continuation of enterprise activities. This is done through introduction of the “early warning” mechanism, a variety of restructuring agreements (discussed under the previous section), and further consolidation of the rules on reorganization (concordato) as the main tool for business continuity in insolvency.

37. Before the new law comes into force, further revisions are possible. Most of the provisions of the new law will come into force within 18 months of the adoption of the law, i.e., in August 2020, with only a few changes to corporate law having an immediate effect.29 Meanwhile the government has been granted the mandate to propose changes, where necessary, within the scope of the original reform principles as laid out in the law outlining the principles for the current reform of the Italian insolvency regime.30 This period of vacatio legis is intentional and is meant to allow time for the adoption of the necessary implementing measures and creation of the supportive infrastructure (including establishment of the OCRI, adoption of the financial distress indicators and framework for insolvency administrators etc.). This period also allows the stakeholders to familiarize themselves with the new law and provides opportunity for potential suggestions for fine-tuning.

38. The new law focuses on business continuity, bringing changes to the reorganization procedure (“concordato”). The current framework already provides for a modern type of reorganization process, generally in line with best practices. It includes comprehensive rules for post-petition financing and offers a number of restructuring tools, including debt-to-equity swaps. It also allows the sale of the business as a going concern subject to certain conditions. The intention of the new law is to further ensure the use of the institute of “concordato” as a tool for restructuring businesses and preserving business continuity rather than as a creditors-managed liquidation, as in past practice. The new law provides that assets-sale reorganizations within the reorganization framework are admitted only under limited/exceptional circumstances, making liquidation-type of reorganization almost impossible.31 The approval of a plan requires the vote in favor of 51 percent of outstanding credit. The law also introduces the additional requirement that in cases where one creditor holds the majority of votes the plan needs to be also approved by the majority of creditors in number. Another novelty introduced by the reform is the requirement of the judge’s assessment of the economic feasibility of the restructuring plan, albeit based on an expert opinion, in addition to the current requirement of assessing legal requirements. The court approval of concordato can be a long process and take months to complete.

39. The liquidation process remains largely unaltered: it is a lengthy process where numerous creditor priorities reduce the recovery of unsecured claims. The new law does not make major changes in the liquidation stage of insolvency process. However, it further fosters development of informatization of the processes which is an important advantage.32 The new law also strengthens court controls over the asset sales. It modifies the current system of the general authorization to liquidate assets granted to the insolvency administrator and introduces the requirement of court approval of each single act of liquidation. Such additional level of court control over each liquidation act risks prolonging the liquidation stage. Furthermore, the current system of multiple priorities dispersed in different laws complicates the framework, often compromises the recovery options for the creditors and contributes to uncertainty. The original intention of the reform to consolidate and reduce the list of priorities has not materialized and the law contains no changes in this regard. Further consideration of this reform is warranted.

40. Among the achievements of the new law are the provisions on coordination of insolvency procedures of companies belonging to the same group. While coordination is not part of the current legislation, it has been already applied by Italian courts and explicit new law provisions to this regard will further facilitate and spread this practice. The new law provides that companies of the same group are allowed to file before the same court (and regardless of the location of their registered office as is currently the case) a single petition for the approval of a debt restructuring agreement or a single concordato preventivo plan and proposal, or a petition for the commencement of judicial liquidation proceedings. This will not lead to the consolidation of assets and liabilities of the different companies belonging to the same group.

41. The new insolvency law of 2019 brings no changes to the special regime for large enterprises which remain to be partially exempt from the general insolvency regime. A unique feature of the Italian system is the special insolvency regime for large enterprises known as “extraordinary administration” (amministrazione straordinaria).33 The regime applies to large distressed enterprises that satisfy certain requirements in terms of number of employees and size of debt. The special regime is meant not to orderly liquidate the company but rather to ensure the temporary continuation of the distressed large companies while allowing them to safeguard employment and production levels during the insolvency procedure. The regime therefore accords special protection to broader social and employment preservation considerations over creditor rights. If the business cannot be restructured, it will be eventually subject to liquidation. This is contrary to the international standard on insolvency. More generally the existence of the special regime likely interferes with general economic efficiency compromising investment and competitiveness.34

42. Another positive objective of the reform is to strengthen the institutional framework for insolvency. The law establishes the creation of a register of insolvency professionals at the Ministry of Justice and provides for further adoption of the qualification requirements and continuous education/training. The new rules–when adopted in the coming year–are expected to increase the professionalism and accountability of insolvency administrators incentivizing good practices. At the same time, the insolvency administrators’ powers will be augmented allowing them to bring actions against shareholders and directors of the company responsible for its economic distress.

43. The lack of judicial specialization in insolvency has represented a continuous challenge for Italy and has not been appropriately addressed under the latest reform. In addition to the general judicial system’s endemic delays, the problem of the lack of specialization in insolvency has been commonly signaled as affecting the operation of the insolvency system. The current law does not address this issue and the initial considerations on this matter have not been taken.35 Apart from the introduction of COMI (i.e., center of main interests of the debtor) as the criterion to determine territorial competence, the only other significant change to jurisdictional competence has been to transfer the jurisdiction over large enterprises and groups of enterprises to the courts with specialized sections in enterprise matters established in large cities, which usually include specialized insolvency judges.36

44. As the new law will strengthen the supervisory power of the courts over the insolvency process, the efficiency of the courts will further influence insolvency outcomes. The court will be granted greater powers on confirmation of reorganization plans (by the requirement of the feasibility assessment), liquidation of assets and general supervision of the process. While court supervision provides the necessary safeguards for the parties and the integrity of the process, its efficiency heavily depends on the ability of the judge to make speedy and expert decisions. Without improvements in judicial expertise, speeding up internal court processes and the harmonization of court practices across the system, the new additional powers of the court allows for risks of further delays to the process.

Recommendations

  • Reinforce the new insolvency law with the revision of creditor priorities and further streamlining of in-court processes. Consolidate the system of multiple special priorities under the insolvency law, reducing them whenever possible. Revisit the new requirement for judge’s approval of each liquidation act and return to the current rule providing the administrator with a general mandate to liquidate the estate.

  • Bring the extraordinary administration regime for large enterprise into the general insolvency regime. Large enterprises and its creditors should be able to benefit from the general insolvency rules both when it comes to restructuring and liquidation. The ability to renegotiate all contracts, including labor contracts, during the restructuring process is an important element for effective corporate restructuring. Secured creditors should also benefit from the general safeguards limiting the stay on creditors’ actions. Judicial control over the process is key for impartial and efficient outcomes of the restructuring or liquidation of distressed businesses.

  • Within concordat o framework allow the expedited court approval of reorganization plans already agreed by the necessary majority of creditors (fast track pre-pack). Fast track plans could be subject to additional supporting documentation requirements to facilitate and speed up the court’s review.

  • Increase the competences of the enterprise courts allowing concentration of commercial and insolvency disputes and promoting judicial specialization. Considering the general delays of the court system and the general complexity and time of insolvency processes, these should be dealt by judges with special expertise on the matter. Further consideration of this issue is necessary, including as per the proposals in the law outlining the principles for the insolvency reform.37 The knowledge and expertise of judges in insolvency matters should be progressively increased.

  • Complete and implement the framework for the regulation and supervision of insolvency administrators. The new rules, envisaged by the new insolvency law, should ensure the continuous build-up of expertise and the accountability of insolvency administrators.

1

The authors of this technical note are Dermot Monaghan and Natalia Stetsenko.

2

Data on projected reductions do not include the NPL plans presented by BCC merged in the cooperative banking groups

3

In Italy, NPLs classified into three categories: bad loans, UTP, and past due. The first comprises loans to gone concern debtors, whose cure rates are close to zero. The second includes loans to going concern debtors, having a non-negligible probability to return to the performing status but in serious financial difficulties, possibly past-due over 90 days. The third comprises exposures that are neither bad loans nor UTP but are past-due over 90 days.

4

EU regulation 2019/630 enacted on 17 April 2019 and applies progressive provisioning to 100 percent within 3 years for new unsecured exposures, 7 years for those secured with movable collateral, and 9 years for immovable collateral. The SSM’s August 2019 communication expects that by 2021, banks should be 100 percent provisioned after 3 years for unsecured exposures, 7 years for loans secured by moveable collateral and 9 years for loans secured by immovable collateral for exposures that become non-performing from April 2018. In addition, in January 2019 the SSM bilaterally communicated to banks with high non-performing exposures (NPEs) an expectation that the existing stock of NPEs will be fully provisioned within 5–6 years.

5

There is a small proportion of loans that are classified as past-due, but not as unlikely to pay.

6

While NPL stock was reduced significantly in 2019, the coverage ratio decreased only marginally.

7

The analysis was based on June 2018 bank data and bad loan value recovery data up to end-2017. It assumed that over the lifetime of the UTP loan portfolio, 50 percent of exposure will become bad loans, 20 percent will return to performing status, and the rest (30 percent) will be closed in banks’ books as UTP. This analysis also assumes bad loans are disposed of first and accounts for provisions on existing UTPs that are assumed to return to performing.

8

Data do not include the NPL plans presented by BCC merged in the cooperative banking groups.

9

According to the BdI, the NPLs of these banks represent around 90 percent of the NPLs of Italian LSIs (other than BCCs).

10

High NPLs as described by the 2018 EBA guidance are those greater than 5 percent of gross customer loans.

11

The EU draft regulation, expected to be enacted in April/May 2019, envisages progressive provisioning to 100 percent within three years for new unsecured exposures, seven years for those secured with movable collateral, and 9 years for immovable collateral. The SSM’s 2018 addendum guidance expects that by 2021, banks should be 100 percent provisioned after two years for unsecured exposures and seven years for secured exposures that become non-performing from April 2018. In addition, in January 2019 the SSM bilaterally communicated to banks with high NPEs an expectation that the existing stock of NPEs will be fully provisioned within five to six years.

12

Issues related to enforcement of household and consumer debt framework are outside of the scope of this assessment.

13

The reforms, inter alia, modernized the corporate reorganization framework (“concordato”) (2005–2006); enhanced the creditors’ decision power in the insolvency process (2009); facilitated the use of out-of-court restructuring agreements including through provisions for bridge financing, the temporary stay of creditor actions, and protection against avoidance actions (2012); further strengthened reorganization procedures and sought the reduction of procedural delays (2015).

14

The Bank of Italy looked at the correlation between the time of recovery and the price investors are willing to pay for a bad loan, and estimated that shortening the recovery time by a year could increase the price by 4.6 percent of gross book value. Under such estimates and with the assumption of 20 percent IRR, the price of a bad loan progressively decreases from 36.2 percent to 12.9 percent of gross book value by the 6th year of recovery. See the BdI, Notes on Financial Stability and Supervision, “What’s the value of NPLs”, April 2016.

15

See the BdI, Notes on Financial Stability and Supervision, “Bad loan recovery rates in 2017”, N13, December 2018.

16

The estimate unattended needs of the judicial system for in-court personnel (not including judiciary staff) stood at 24 percent and 20 percent between 2017 and 2018 according to a recent statement of the Minister of Justice.

17

The procedural deadlines for submission of cadastral documentation were shortened from 120 days to 60 days and for appraisal from 120 to 90 days; the auction sale now allows to reduce the price by 25 percent from appraisal value after each failed auction and up to 50 percent of appraisal value after the fourth failed auction.

18

According to the initial estimates of the BdI, based on the sample study following the reform to the foreclosure process by DL. 83/2015 and 59/2016, the foreclosure time for real estate has been shortened in 2017 on average between 10–15 percent within pre-sale procedures and up to 43 percent for the sale procedure. See, BdI, Questioni di Economia e Finanza, Real estate foreclosure procedures and the impact of the reforms, July 2018. See http://www.bancaditalia.it/pubblicazioni/qef/2018–0448/QEF_448_18.pdf ).

19

Portale delle vendite pubblichehttps://pvp.giustizia.it/pvp/. Under Art. 569, co. 4 of the civil procedure code, as amended by the 2016 reform, auctions are to be conducted through electronic systems with the aim of facilitating access and increasing participation.

20

For example, property descriptions are found through complicated technical documentation, without proper exposure of the assets, pictures are scarce and do not give sufficient visibility to the property on sale.

21

Particularly, smaller courts where a single (and sometime newly appointed) judge handles all legal matters (including criminal cases) under the court’s territorial jurisdiction are generally most affected by the lack of expertise in more complex commercial matters.

22

Italian law has recovered the Roman law construction of “patto Marciano”, which is an agreement between the creditor and the debtor that establishes the creditor’s right to repossess the assets in case of default. As a fundamental safeguard for the debtor the law provides for an independent and professional appraisal of collateral value and in case such value exceeds the outstanding debt, the surplus is returned to the borrower.

23

The MOU states that the repossession of collateral under the Marciano mechanism does not imply an automatic discharge from the full amount of debt where the value of collateral is not sufficient to cover such full amount. It also provides that, according to the credit agreement, the creditor may be allowed to opt-out from the repossession procedure if the evaluation of the real estate provided by the valuator is deemed unsatisfactory. The MOU states that the out-of-court sale of collateral (and not only repossession) could be a legitimate form of out-of-court enforcement if contemplated in the loan agreement. This allows the bank to hold a direct sale via a specialized agent without the need to acquire the assets first. It is unclear to what extent the sale can be carried out without the explicit law provision allowing such sale.

24

The law provides for Marciano enforcement mechanism with regard to consumers too but with some significant differences in the rules, compared to enterprises. For consumers the law explicitly allows that the Marciano clause could include not only repossession but also a direct sale of collateral. However, the secondary regulation on such sale is yet to be adopted. On the other hand, the application of Marciano mechanism always leads to debt discharge in the context of consumer credit enforcement while for corporate debtors the discharge needs to be explicitly agreed by the parties.

25

Under Art 15 of the new insolvency law three categories of public creditors (tax Administration, Social security Agency, debt collecting agency) are under obligation of starting the alert procedure via OCRI if certain law provided thresholds of taxes and duties remain overdue for a certain period of time, and the debtor fails to remedy the situation or initiate restructuring solutions within 90 days of the receipt of the notification to this regard from the relevant public creditor.

26

OCRI are crisis advisory organizations, which are yet to be created before the law enters into full force in August 2020. During the restructuring negotiation, the debtor may also apply to the court for the adoption of protective measures (e.g., a moratorium).

27

See Jose Garrido, Insolvency and Enforcement Reforms in Italy, IMF WP/16/134 (2016).

28

The single procedure applies for filing of restructuring agreements (accordi di ristrutturazioni), reorganization (concordato) and liquidation.

29

With an immediate effect the new insolvency law has introduced a number of changes into corporate law aimed at strengthening the companies’ internal control mechanisms for crisis detection. In particular, the law provides for an obligation on the directors to put in place and maintain an adequate organizational, administrative and accounting structure aimed at a prompt detection of financial distress. Furthermore, the requirement for appointing the internal control bodies has been extended to a greater number of companies. In addition, the new law also increased the financial liability for the directors in case of economic damage.

30

Law 19/10/2017, n. 155, Gazette 30/10/2017.

31

Under the 2019 insolvency law the liquidation-type of reorganization plan will be allowed only when they include proposals for asset sales where third parties add additional resources to the insolvency estate, increasing the potential recovery of unsecured creditors by at least 10 percent (compared to the judicial liquidation scenario) with a recovery rate of at least 20 percent of their debt.

32

In consideration of the long insolvency processes the law lengthens the current maximum time of liquidation from the current 2 years (extendable) to 5 years. This seems to be no recognition of the actual average time for the insolvency in Italy. At the same time, the new law requires to organize the first liquidation sale within a year of the opening of the liquidation procedure.

33

The special regime is characterized by two types of extraordinary administration procedures. These are generally referred to as “Prodi-bis” and “Marzano” procedures (for large enterprises and very large enterprises, respectively). Both procedures are largely focused on preserving economic activities and maintaining employment’s levels in large enterprises rather than maximization of creditors’ interests as in normal insolvency processes. The procedures represent a hybrid of a judicial and an administrative procedure and assign a prominent role to the Ministry of Economic Development and the commissioners appointed by the Ministry who perform functions similar to those of insolvency administrators.

34

See Nazim Belhocine, Daniel Garcia-Macia, and José Garrido, The Insolvency Regime for Large Enterprises in Italy: An Economic and Legal Assessment, IMF WP 18/218 (2018).

35

The law outlining the principles for the reform of the Italian insolvency regime (law 19/10/2017, n. 155, Gazette 30/10/2017), which served as the basis for the current reform, proposed to concentrate commercial insolvency cases in the enterprise courts while preserving the competence of courts in cities where there are no enterprise courts. In this regard, the law specified the following criteria to assign the competence to courts of cities where there are no enterprise courts: based on the series of criteria which included judicial capacity considerations, past experience in handling insolvency cases and numbers of expertise subject to court jurisdiction.

36

As per Art 1 of the legislative decree n 168 of June 28, 2003, these are the courts and appeal courts of Bari, Bologna, Catania, Firenze, Genova, Milano, Napoli, Palermo, Roma, Torino, Trieste and Venezia.

37

Law 19/10/2017, n. 155, Gazette 30/10/2017.

Italy: Financial Sector Assessment Program-Technical Note-Tackling Non-Performing Assets
Author: International Monetary Fund. Monetary and Capital Markets Department