NPLs are a dominant problem for banks in the euro area as in some countries almost one quarter of loans are not serviced. NPLs represent a real challenge for bank profitability and financial stability. In addition, they constrain credit expansion and delay economic recovery.
Despite some recent progress, slow growth and persistent unemployment as well as low investment interest due to asymmetric information and a wide bid-ask price wedge, make extremely difficult the cleaning of banks’ balance sheets.
A series of options have been suggested with a view to improving conditions in the European NPL market and reinforcing investor confidence respecting at the same time state aid rules. Public intervention measures, such as asset management companies and other co-investment strategies are deemed necessary in order to increase market efficiency and create a virtuous circle of reductions in NPLs and increases in investment and growth much needed in the euro area.
1. A high stock of NPLs
Non-performing loans (NPLs) are a dominant problem for euro-area banks, as they exceed 6.6% of all loans in 2016 compared to 1.5% in the US. Total NPLs reach 1 trillion euro, while uncovered (after considering provisions) NPLs are more than six times the annual profits of EU banks (Enria, 2016). Especially for a group of six countries (Cyprus, Greece, Italy, Ireland, Portugal and Slovenia) NPLs reach 22.8% and represent a real challenge for bank profitability. Moreover, such prominent and persistent differences in NPLs can be interpreted as a clear sign of fragmentation in the euro-area banking market. In a recent report ECB (2017b) states explicitly that in some euro-area regions the prospects of banks’ profitability continue to be depressed due to the large stocks of NPLs in their balance sheets. In addition to being a drag on profitability, NPLs constrain credit expansion, endanger financial stability and delay economic growth (Constancio, 2017).
NPLs are also closely related to the problem of debt overhang which acts as a disincentive for highly leveraged firms to ask for credit in order to finance new profitable projects and, consequently, it suppresses demand for corporate investment. In addition, non-viable firms may be kept alive by already committed banks while at the same time viable firms suffer from lack of funding and unhealthy competition. Thus, recovery is further delayed (Demertzis and Lehmann, 2017).
But what creates NPLs? They are caused mainly by (a) macro-economic factors (which characterize recessions) such as lack of growth, increasing unemployment, high interest rate margins, reductions in disposable income and increasing tax burden, and (b) bank-related factors such as management skills and risk preferences. Moral hazard may also play a significant role (Anastasiou, Louri and Tsionas, 2016). The recent financial crisis in Europe combined most of these factors and created conditions of heavy systemic stress in the banking sector which led to the current large stock of NPLs. For as long as slow growth and high unemployment persist, especially in some countries, NPLs will continue being a serious problem for their economies and for the entire euro area because of related spillovers. Even if macro-economic conditions improve and bank management becomes more efficient the current stock of NPLs is so high that it will need a long period of time to reach acceptable levels.
2. Resolution strategies
To reduce the NPL stock faster and more effectively public intervention measures are required in the euro area. Such measures should help removing the impaired assets from the banks’ balance sheets swiftly and without triggering requirements for capital injections which will not be easily manageable. There can be two ways both of which should be complemented by appropriate reforms in the legal framework facilitating debt enforcement:
a) The banks enhancing their efforts to manage NPLs on their own through internal NPL workout and external servicing. Enhanced supervisory guidance (ECB, 2017a) can be helpful in setting ambitious targets and restructuring plans but it is unlikely to be sufficient in current circumstances. And
b) The banks transferring impaired assets to a third party, such as outright sales to investors, or to a special purpose securitization vehicle or to an asset management company (AMC). Since fire sales are to be avoided the most important questions related to transferring NPLs are through which mechanism to proceed and at what transfer price. The difference between the net book value (nominal book value minus provisions) of the impaired asset and its transfer price is the loss incurred by the bank at the time of the transfer. Since markets for impaired assets are rather illiquid at present due to lack of symmetric information about their quality and legal uncertainties about their recovery time and process, there is a first-mover disadvantage for banks selling NPLs which may lead to higher losses.
In order to face the challenge and help banks restore their intermediation function and support the real economy again a comprehensive strategy using new and radical tools has to be designed and implemented. Otherwise, the existence of such legacy assets will create a doom loop, whereby slow growth increases NPLs and NPLs obstruct further growth due to lack of finance. In addition, the debt overhang, which often forces business to deleverage, represses investment and though hysteresis effects further delays recovery.
Aiyar et al. (2017) argue that a comprehensive approach to the NPL problem comprises three pillars: (1)intensified oversight by the supervisor imposing detailed schedules of gradual NPL reduction and more conservative provisioning; (2)enhanced judicial procedures as well as an out-of-court restructuring framework; and (3)developing markets for NPLs by reducing asymmetries in information, improving the available infrastructure and establishing asset management companies (AMCs) to exploit economies of scale, scope and specialization. The Swedish experience of the early 1990s where NPLs had reached 11% and bank lending contracted by 26% between 1990 and 1995 is an excellent example of the potential an AMC has to clean the market and help the real economy start growing again. It is useful to underline that large part of the Swedish AMC’s success was due to efficient focusing on homogenous types of loans (Demertzis and Lehmann, 2017).
For an effective reduction in the 1 trillion euros of NPLs that European banks currently hold in their balance sheets conditions in the market for NPLs have to be improved. Constancio (2017) distinguishes the market impediments from the demand side as information asymmetry, uncertain debt enforcement, transfer restrictions and complicated licensing procedures for those interested in acquiring delinquent debt. On the supply side are the banks which would like to transfer NPLs close to their book value and are unwilling to realize losses for which provisions are not enough. Such structural inefficiencies may drive a wide wedge between bid-ask prices which may exceed 40% in some cases even if the loans are fully backed with collateral.
An important strategy is thus, to reduce all those impediments in order to facilitate the market mechanism and create a liquid and smoothly functioning secondary market for NPLs. Structural reforms which facilitate symmetric information and transparency to all participants e.g. through platforms and clearing houses improving the capacity of the judicial system as well as the out-of-court resolution process, increasing the range of restructuring options and reducing their enforcement cost can be very helpful. The creation of an AMC may also be helpful in initiating market transactions.
Although the success of the Swedish and Asian experiences has stressed the potential of an AMC in clearing the market smoothly and relatively fast, and such an entity has already been used in Spain and Ireland with some positive effects, a state backed AMC is considered as state aid in Europe and following the BRRD (implemented as from 1st Jan 2016) it is allowed only under certain conditions.
In addition, there is the problem of the appropriate price. Avgouleas and Goodhart (2017) explain that a major disincentive for banks to clean up the pile of NPLs is the price of sale of NPLs which, if it is below the net book value, will generate a capital write off. European banks already have low profitability making it difficult to absorb further losses. Recording a serious loss of capital may trigger the bail-in process (introduced by the BRRD) with dire consequences for some of the bank’s stakeholders. A series of such bail-ins could destabilize the banking sector of a country with further systemic consequences (Bruno, Lusignani & Onado, 2017). Any mechanism which can help avoiding such undesired developments by correcting the transfer price is worth trying.
An AMC extends a visible and credible commitment of a fiscal backstop which puts a cap to bank losses. On top, AMCs can enjoy synergies especially when dealing with homogeneous types of assets (e.g. commercial real estate or corporate loans) and hire specialized skilled professionals to restructure the delinquent assets and negotiate appropriate prices as well as debt/equity conversions. They can also be more efficient at securitizing NPLs adding to the liquidity and the depth of the secondary market. On the negative side, AMCs may face uncertainty about the quality of the assets especially if transparency is limited. Consequently, the choice of the appropriate valuation method is very important. A critical issue is also the question of loss absorption.
Avgouleas and Goodhart (2017) recommend a pan-European AMC (EAMC) as the most effective approach to manage the accumulated delinquent assets in the euro area. The EAMC will be a holding company of national AMCs and will be funded through proportionate contributions by member states. The EAMC will hold a 10% stake in the national AMCs which will be set up as subsidiaries. National banks will hold the remaining 90% of the national AMC’s (their contribution depending on their share of NPLs) which will operate under the same conditions of governance, transparency and disclosure. Strong uniform governance links will connect the EAMC with its subsidiaries while redistributive outcomes are to be strictly avoided.
The price at which the impaired assets will be transferred can be a combination of their book value excluding provisions, their real (long-term) economic value and their market value. Thus, a more balanced valuation will take into account the potential rise in market prices once the economy rebounds. Objectivity can be assured by holding auctions for similar assets and asking the EIB to conduct the real economic value estimations. Profit and loss agreements between the banks and the AMCs can be accompanied by an ESM guarantee within the framework of the ‘precautionary recapitalization’ process and a respective conditionality on their business plans. Thus, private bondholders of the AMCs will be less exposed to losses and private interest in financing the project can be broader. Reporting of the EAMC to the SSM, the EC and the ESM will be on semi-annual basis.
The EAMC has a comparative advantage to establish a centralized platform where information sharing and direct sales of assets can take place, facilitating the function of the secondary NPL market and boosting liquidity. Transparency of outcomes will improve accountability and help remove moral hazard. Also, based on its market information, the EAMC will be able to determine the optimal speed of NPL resolution and achieve the most efficient outcome. Avgouleas and Goodhart (2017) also stress how efficient the EAMC-national AMCs structure will be in attracting new private funding both for the AMCs and the national banks as well as freeing up capital for new lending and growth.
In the same vein, Haben and Quagliarelo (2017) argue in favor of a European AMC which could enjoy some public (capital) support within the BRRD framework. The aim is to improve clarity of information, reduce funding costs and lead to higher operational efficiency. After calculating a state aid envelope for each bank based on stress test results, part of this theoretical state aid (allowed for precautionary recapitalization) can be used to fund the transfer of NPLs to the AMC at their real economic value. Since real economic value will be lower than the book value, capital write offs may follow and some bail in may be needed. Furthermore, an ex post clawback provision may require banks to bear part of the loss if the real economic value is not achieved after a specified period (e.g. three years). Such a clause discourages moral hazard.
For not causing further uncertainty to the banks impaired assets will be irrevocably transferred at the point of sale. The clawback could have the form of equity warrants to the government which are to be triggered if the final sale price is lower than the transfer price (assumed to be the real economic value) and state aid will thus be introduced with full conditionality. Such a provision creates a further incentive to the banks to agree on an objective real economic value and have skin in the game to the end. The proposal avoids burden sharing since it will be each national government stepping in for a capital injection needed by a national bank. Haben and Quagliarelo (2017) also argue that their blueprint leading to a more efficient secondary market for NPLs has the potential to uplift supervisory expectations with regard to management and reduction of NPLs (ECB, 2017a).
Fell, Moldovan and O’Brien (2017) elaborate on the role of the national AMCs bridging the time between the current period of high risk and depressed prices of the distressed assets and the future improvement of the economy which can lead to maximizing recovery values. Following transfer of the depressed assets, banks enjoy an improvement in funding and capital costs. Dedicated legislation and a finite lifetime are essential prerequisites for the success of an AMC. The legislation should introduce the appropriate governance, objectives and accountability rules. Outsourcing, (e.g. of legal services, collections, etc.) should be used in order to keep the AMC structure light and flexible.
Regulatory constraints as imposed by the BRRD and the state aid directives of the EC must be respected. The capital structure should not allow the government to own more than 50% thus avoiding political interference and control. Government guaranteed senior bonds can be provided as capital which (if they meet eligibility criteria) can be further deposited for funding from the Eurosystem or even from the interbank market. Participation in the AMC should be incentivized, while the perimeter of the assets included should be clearly defined. Homogeneity of assets as well as a minimum size threshold (i.e. only significant exposures) are necessary conditions for optimizing the function of the AMC and achieving a high recovery value. The asset valuation should be as objective as possible, close to the real economic value and conducted by an independent authority.
2.2. Securitization, co-investments and other measures
Apart from an AMC, Fell, Moldovan and O’Brien (2017) propose three additional approaches to help the swift reduction of NPLs in the euro area:
a) Platforms which can be used as a central hub for providing due diligence type of information and facilitating the exchange of NPLs and/or even selling such assets. The idea is to reduce information asymmetries and other investor costs.
b) Securitization instruments or other co-investment structures which promote risk-sharing and improve recovery value. And
c) Liquidation frameworks for those assets whose value is unlikely to be recovered. A ‘central liquidator’ may be designed as a public entity the purpose of which is following the member state’s liquidation legislation to recover part of the value and return it to the legal recipients. The proceeds of the liquidator depend a lot on the general macroeconomic conditions and especially the recovery speed of the economies under stress.
Galand, Dutillieux and Vallyon, (2017) elaborate on the state aid which is considered to be the difference between the market price of an asset and the transfer price paid e.g. by a state-backed AMC. A higher transfer price may be necessary to allow the bank to remove the impaired asset from its balance sheet without causing a major capital write off. For the related state aid to be considered compatible with the internal market a cap has been suggested at the real economic value of the purchased asset. Real economic value is the present value of the future net cash flows from the asset and as macroeconomic conditions improve the market price will converge to the real value. The discount factor used to find the present value should take into account the time needed and the risks taken until the value is realized.
The efficient operation of platforms could help retrieve such information easier and faster. The recent experience in Europe shows that on average real economic value is 10-15% higher than the market price and that it has been a good proxy of the realized proceeds from the final sale of the impaired assets. With reference to the state aid, banks should be protected only from unexpected losses due to extraordinary circumstances. It is obvious that accurate valuation is very important but at the same time it is very difficult as there is no liquid market for NPLs. The EC trying to combine all the available information so as to provide accurate valuations has often been criticized for delays which could be avoided if reliable platforms were in operation.
External servicing companies could also be facilitated by platform-provided information. Servicers, as they undertake the management of a sub-set of NPLs, can be used to relieve some of the pressure banks receive from their clients and reduce moral hazard. Banks can outsource part of the management of NPLs while keeping them on their balance sheets. Outsourcing servicing through e.g. an SLA agreement can be a cost-efficient solution which may also help banks redistribute their staff to more profitable activities.
Another type of impaired asset relief measure suggested by Galand, Dutillieux and Vallyon (2017) is asset guarantees by the State which commits to bear a remaining part of the loss after the bank incurs an agreed first part. The guarantee kicks in only if the loss exceeds the “attachment point”. For this relief the bank has to pay a guarantee fee to the State, which has to be comparable to what a market economy insurer would ask. If the fee is high enough and the probability of the attachment point reached low, then the measure can be considered as aid-free. The guarantee can thus have two or three tranches (senior, mezzanine and junior) if it is agreed that the bank pays a certain percentage of the loss above the attachment point. The idea is again that the bank is protected against losses due to extraordinary and not foreseeable circumstances.
Since both real economic value transfers and asset guarantees are a form of restructuring aid, they have to be accompanied by specific conditions showing that the bank follows a comprehensive restructuring plan, and at the same time shares the expected part of the losses incurred due to its bad loans. It is imperative to avoid distorting competition in the market as a consequence of the relief measures extended only to some banks. Hence, divestment of subsidiaries, of branches or of non-core assets may be required and restrictions on management remuneration, pricing policies, marketing strategies can be imposed. Opening up the market and facilitating new entry can also be pursued.
Bruno, Lusignani and Onado (2017) propose a securitization scheme which could increase bid prices and provide a significant cleaning of impaired balance sheets. They correctly point at the main problem which is the appropriate transfer price. On the one side banks are reluctant to dispose of bad loans at prices not covering the net book value and on the other market uncertainty raises the risk premium demanded by investors and thus lowers the transfer price. Asymmetric information creates a ‘lemons market’ where trade eventually stops and some kind of risk absorber will be needed for uncertainty to recede and trade to start again. Thus, an official intervention could facilitate exchange through e.g. the creation of risk-sharing investment vehicles.
According to the blueprint by Bruno, Lusignani and Onado (2017) NPLs can be sold to an SPV which can then issue different tranches of securities. The tranches will provide different risk-return combinations and will thus be appealing to different categories of investors from risk-averse (buying the senior tranche) to risk-lovers (buying the junior tranche). Using euro-area data the authors estimate the benefit from two alternative schemes based on different assumptions about capital costs and transfer prices. Their exercise shows that in either case such securitization techniques can seriously reduce the ‘big mountain’ of European NPLs. Banks will record an immediate, contained loss due to the lower transfer prices but there must be enough buffers to absorb it. The subscribers of the riskier junior tranches may need some public support in the form of a guarantee.
ECB (2017b) also promotes co-investment (private-public) strategies in order to bridge the bid-ask spread putting forward two different instruments relating to either securitization or direct sale of impaired assets. The central idea put forward by Fell, Moldovan and O’Brien (2017) is for the state to signal its aligned interests with banks and investors. The first instrument is a state guarantee of junior tranches of NPL securitization (JGS) of e.g. up to 50% of the losses in return for any better performance of actual recoveries. Thus, the state is exposed to the same risk as the investor, while the investor can choose the level of protection he desires. It is this freedom of choice of the degree of protection which can attract a larger group of investors opening up the market and eventually increasing the price paid to the bank selling the NPLs.
The second co-investment (forward purchase) scheme (FPS) suggested by Fell, Moldovan and O’Brien (2017) is a type of loan provided by the state to the investor in order to finance part of the purchase price of the transferred NPLs. The purchase price is actually a forward price at the maturity of the scheme (e.g. in 5 or 7 years). Since the bid price at the time of the transaction is lower, the state extends a loan to the buyer of the NPLs equivalent to the ‘forward premium’. In order to contain the risk to the state the buyer has to have as guarantor an investment grade institution. With the support of the FPS the transfer price of the NPLs is higher and thus closer to the ‘ask’ price by the bank, improving the functioning of the NPL market.
The European NPL problem has reached dramatic proportions as Bruno, Lusignani and Onado (2017) stress and is delaying economic recovery. A European solution or, rather, a combination of resolution tools and strategies has to be put in place in order to face the challenge.Despite some recent progress, slow growth and persistent unemployment (especially in some euro-are countries) as well as low investment interest due to asymmetric information and a wide bid-ask price wedge, make extremely difficult the swift cleaning of the banks’ balance sheets.A series of options have been suggested by researchers with a view to improving conditions in the NPL market and reinforcing investor confidence respecting at the same time state aid rules. Public intervention measures are deemed necessary in order to increase market efficiency and create a virtuous circle of reductions in NPLs and increases in investment and growth much needed in the euro area.
Aiyar, S., Bergthaler, W., Garrido, J.M., Ilyna, A., Jobst, A., Kang, K., Kovtun, D., Monaghan, D., and Moretti, M. (2015). A Strategy for Resolving Europe’s Problem Loans. IMF Discussion Note.
Anastasiou, D., Louri, H., and Tsionas, M. (2016). Determinants of non-performing loans: Evidence form euro-area countries. Finance Research Letters, 18, 116-119.
Avgouleas, E., and Goodhart, C. (2017). Utilizing AMCs to tackle the eurozone’s legacy non-performing loans. European Economy – Banks, Regulation and the Real Sector, this issue.
Bruno, B., Lusignani, G., and Onado, M (2017). A securitization scheme for resolving Europe’s problem loans. Mimeo.
Constancio, V. (2017). Resolving Europe’s NPL burden: challenges and benefits, speech, Brussels 3 February 2017.
ECB (2017a). Guidance to banks on non-performing loans, March 2017.
ECB (2017b). Financial Stability Report, May 2017.
Enria, A. (2016). Financial markets 2.0-R(evolution): Rebooting the European banking sector, speech, Vienna 5 October 2016.
Fell, J., Moldovan, C., and O’Brien, E. (2017). Resolving non-performing loans: A role for securitization and other financial structures, in ECB (2017b), 158-174.
Fell, J., Grodjicki, M., Martin, R., and O’Brien, E. (2017). Resolving large NPL stocks in Europe – A role for systemic AMCs in a comprehensive solution (ECB)
Galand, C., Dutillieux W., and Vallyon, E. (2017). Non-performing loans and state aid rules (EC).
Haben, P., and Quagliariello, M. (2017). Why the EU needs an asset management company (EBA).