Ignazio Visco: Italy's economy and banking sector in the aftermath of "Brexit"

Address by Mr Ignazio Visco, Governor of the Bank of Italy, at the 56th Italian Banking Association Annual Meeting, Rome, 8 July 2016.

The views expressed in this speech are those of the speaker and not the view of the BIS.

The referendum in the United Kingdom

The predominance of votes against the UK remaining in the European Union in the referendum of 23 June has created an unprecedented situation in the history of European integration, whose longer-term implications are difficult to predict. The impact of the referendum result on the foreign exchange and financial markets was mitigated by the actions taken by the monetary authorities, though the tensions only partly subsided in the days that followed. After losing about 6 per cent of its value in the 12 months prior to the referendum, since 23 June the pound sterling has depreciated by 11 per cent against the currencies of its main trading partners and by 13 per cent against the dollar, hitting a 30-year low. The euro has appreciated by 11 per cent against the pound but has depreciated against the dollar, the yen and the yuan; in effective terms it has remained substantially unchanged.

The impact on the interest rate differentials between government securities in the euro area has been limited, thanks above all to the stabilizing effect of the Eurosystem's asset purchase programme: the yield on ten-year Italian government bonds (BTPs) has fallen by 16 basis points, the differential with respect to the corresponding German Bund has widened by 11 basis points. Yet share prices have fallen sharply throughout the euro area, especially in Italy, and volatility has surged. Increased risk aversion among investors has mainly hit securities in the sectors most exposed to a slowdown in growth, among which the banking sector, already weighed down by concerns over balance sheets; the fall in the stock prices of Italian banks was steeper than the euro- area average, with shares tumbling by 29 and 21 per cent respectively.

The long-term repercussions of the referendum result on the UK and euro-area economies will depend on the new arrangements for trade and finance resulting from complex negotiations whose outcome is uncertain; in the short and medium term the scale of the effects will depend on the speed and clarity of purpose with which the political authorities respond to this new shock.

The changes in interest and exchange rates observed so far do not significantly affect Italy's prospects for growth. There could be some limited consequences connected to trade links in the event of a sharp slowdown of the British economy: a 10 per cent fall in UK imports, close to the upper limit of the leading analysts' evaluations, would shave around one quarter of a percentage point off Italy's GDP in the three-year period 2016-18.

Higher risks to the medium-term outlook for the Italian economy and the euro area could stem from the spread of unfavourable expectations which, as our experience of the crisis years has taught us, can have non-linear and difficult-to-predict effects. A tightening of conditions in the financial markets and credit supply, triggered by investors' flight to safety and by an exacerbation of strains on bank securities, coupled with a general decline in business confidence, would harm investment and growth.

What matters most is the ability to adopt measures, at the European and national level, capable of containing the tensions on financial markets, strengthening the banking system and shoring up confidence. The monetary policy toolbox is well stocked but all the other policies must play their part. We must take care that supervision does not itself become a source of tensions. Recognizing the emergence of risks to financial stability, in the last few days the Bank of England has made use of the flexibility inherent in the rules to reduce the countercyclical capital requirements for banks.

Italian banks

Non-performing exposures and capital adequacy

The fall in Italian banks' share prices reflected, as in other euro-area countries, fears that slower growth could dent banks' profitability and impair their capital. It was, however, compounded by the large stock of non-performing loans (NPLs), a legacy of the recession, and by concerns that current market conditions could make it harder to sell these exposures or raise capital.

Italian banks' NPL problem is serious but manageable; it must be clearly defined and tackled. This is being done, taking account of the need to deliver both swift and cost effective solutions. The reduction in the stock of NPLs recorded since last autumn is an encouraging sign.

The loss of almost 10 percentage points of GDP and about 25 per cent of industrial output during the crisis inevitably had a severe impact on Italian banks' balance sheets and on the quality of their loans. Without the double-dip recession, the gross stock of bad loans, considering only loans to non-financial firms, which exceeded €140 billion at the end of 2015, would have amounted to about €50 billion, or 5 per cent of all loans disbursed, close to pre-crisis levels. The growth of NPLs was strongly influenced by recovery times, which until now have been particularly long in Italy; the measures approved in the last year will shorten them significantly. As the Bank of Italy has emphasized in the past, reducing average recovery times even by just two years would have determined, together with a higher valuation of NPLs, a ratio of bad debt to total loans close to half of what it is today.

Most Italian banks are capable of dealing with the still fragile cyclical conditions, lending to the economy and competing efficiently on the market. A number of assessments made in the last few weeks put the recapitalization needs of the entire Italian banking system at somewhere in the order of tens of billions of euros, based on the assumption that the total stock of bad loans, and possibly even some of the loans that are "unlikely to be repaid", must be sold at once by all banks at a price equal to approximately half the value of the bad loans recorded in banks' balance sheets.

In Italy a loan is classified as "non-performing" based on harmonized criteria established at the European level and published by the EBA in 2013, which are largely aligned with those previously applied in Italy and with international practice. For the valuation of non-performing loans in their balance sheets, Italian banks must comply with international accounting and reporting standards (IAS/IFRS). As they permit discretion we have constantly encouraged banks to adopt prudent valuation policies that take sufficient account of the degree of uncertainty in recovery times and flows. Our supervisory action on provisioning in 2012 and 2013 is a clear example of this.

In the approach traditionally taken in Italy, NPLs can be divided into at least two large and very dissimilar categories based on the degree of difficulty faced by debtors: of the €360 billion worth of gross NPLs outstanding at the end of 2015, bad loans accounted for €210 billion; €150 billion were in loans classified as "unlikely to be repaid", past-due or in breach of overdraft ceilings. These two categories of NPLs obviously have different requirements in terms of coverage and write-downs. Both types of loans must be booked in the balance sheets with due prudence, but not at values corresponding to their immediate liquidation. Loans in the second category may return to the performing loans category; indeed, a significant number did so, even in the extremely difficult conditions of recent years. These loans represent "active" credit relationships; it is very likely that even if they experience a particularly difficult phase, debtors will be able to turn a corner and resume the payment of their debts.

As for bad loans, both the write-downs already made by banks and the underlying collateral must be taken into account. Net of write-downs, the amount of bad loans drops to €87 billion, of which €50 billion backed by collateral worth an estimated €85 billion, while the remainder are backed by personal guarantees estimated at €37 billion or are unsecured. The value of the real-estate collateral in the balance sheets of Italy's main banks was revised downward following the euro-area asset quality review (AQR) of 2014, which was conducted using independent, comprehensive and prudent estimates. The average revision was about 10 per cent, compared with 13 per cent for all the euro-area banks subject to review, confirming that the valuation criteria used by Italian banks were substantially correct. In the period following the AQR, the decline in Italy's real estate prices was moderate; it is also worth recalling that mortgages disbursed by Italian banks have a lower loan-to-value ratio than in the other main European countries.

As recent analyses have shown, the bad loan recovery rates actually observed, including in the last few years, are generally in line with the valuations banks use in their balance sheets. This confirms that not even for bad loans can an accurate assessment of their value be based on the assumption that they will be sold immediately. The market value of a bad loan basically depends on two factors: the investor's profit target and recovery times; higher profit expectations or longer recovery times mean lower prices.

Regarding the first factor, in recent years virtually the only buyers in the market for NPLs have been non-European private equity funds seeking very high returns, i.e. 10 to 20 per cent, well above Italian banks' return on equity, which still averages just under 5 per cent. As for the second factor, the reforms recently introduced to significantly speed up the credit recovery process could have a major impact on the value of bad loans; they must therefore be given the time they need to bear fruit. It will also be some time before the necessary improvements are achieved in the efficiency of Italy's courts, whose performance continues to vary significantly across the country.

The European authorities have also acknowledged on several occasions that solving the problem of NPLs will take time. As the Vice-President of the ECB pointed out just yesterday, it is a lengthy and complex process, the positive effects of which will only materialize over the medium term, beyond the horizon considered in the stress tests. Last year the Single Supervisory Mechanism set up a task force on NPLs. Even the news that a simple request for information had been made in this context triggered considerable turmoil on the financial market at the beginning of this year, highlighting the very sensitive nature of information on supervisory matters and the need to handle it carefully. One of the aims of the task force is to recommend to banks, especially those with high volumes of NPLs, the best practices to adopt in developing an effective strategy to deal with them. Divestment is just one of the options available; other strategies include, as we have often pointed out, more efficient management within the group or outsourcing to a specialized credit recovery service.

The majority of NPLs are booked in the balance sheets of banks that are financially sound notwithstanding the impact of the long and deep recession. At the end of last year, "significant" banks with particularly high levels of NPLs and, among the other banks, those with core tier 1 capital ratios below 10 per cent, jointly held €15 billion worth of bad loans net of write-downs, also backed by collateral and personal guarantees.

For these reasons it is not correct to speak of the problem of NPLs as if it were an emergency affecting the entire banking system. Effective supervision involves assessing the actual situation of each bank using detailed information, conducting robust analyses, and taking account of average credit recovery times.

Essentially, Italian banks have suffered from the effects of the recession; for some of them these have been compounded by weaknesses stemming from their ownership and governance structures, and regrettably, in several cases, by fraudulent conduct. In respect of these banks the supervisory authorities - Italian and, since the end of 2014, European too - have acted within the powers assigned to them; we have given account of the actions taken before the competent institutions and other bodies, publishing detailed reports on our website. The supervisory authorities have called for restructuring plans, which have been drawn up, and are closely monitoring their implementation. It will take time to achieve the objectives; a firming up of the economic recovery with a positive spill-over on banking activity will prove decisive. In some cases, the indispensable recapitalization measures requested by supervisors has been hindered by market difficulties; in others, the recent tensions, which have hit share prices hard, call for resolute action, to promptly signal a reversal of trend, and for possible support measures.

The reform process and the coming months

Major reforms have been launched in recent years to allow Italy's financial system to modernize, in line with the new global and European regulatory framework and with economic, social and technological changes. The long overdue reforms of Italy's cooperative banks ("banche popolari") and mutual banks ("banche di credito cooperativo") have been crucial. For the mutual banks, the reform must be implemented quickly and efficiently: public consultation on the secondary legislation will begin shortly.

The merging of two large cooperative banks announced a few months ago will be an important test of the system's capacity for renewal following the reform; we expect other mergers to follow relatively soon. The recent overhaul of the senior management of a leading bank lays the groundwork for an improvement of its capital position and an increase in profitability, in line with supervisors' and market requirements for global systemically important groups.

Other important reforms were adopted between mid-2013 and the first half of this year to remove tax disincentives that have deterred banks from writing down loans and, as I have already mentioned, to speed up credit recovery procedures. These reforms are paving the way for the development of a large and efficient Italian NPL market.

The state guarantee scheme to securitize bad loans ("Garanzia sulla cartolarizzazione di sofferenze") and the launch of a new survey on bad loans will also contribute to the growth of the market. The Bank of Italy has just introduced this survey to collect and rationalize detailed data on bad loan characteristics, on the type of guarantees used to back them and on the status of recovery procedures. We have indeed found that many banks are making insufficient use of modern data management and processing technologies, particularly with respect to loans handled directly by their legal departments. The difficulty of quickly providing potential investors with an adequate and reliable set of information affects the length of time needed to conclude negotiations and is partly responsible for the substantial discounts on the selling prices of bad loans. The first reports should reach the Bank of Italy soon, before the end of next September.

Important private initiatives undertaken recently, such as the establishment of the Atlante fund and other instruments aimed at supporting ailing banks (including those in the mutual bank sector), seek to help banks successfully navigate this delicate transitional phase. The initiatives come as a response to the new European crisis management framework, whose rigid application restricts States' ability to intervene, even when the aim is to avoid contagion and maintain financial stability. But there is still much to be done to improve banks' industrial strategies.

The stress test of Europe's largest banks, coordinated by the EBA and conducted by the ECB on the euro area's "significant" institutions, has almost been completed. The results will be published at the end of this month. Exercises of this kind are now a standard part of the supervisory toolbox; the stress test results, along with many other quantitative and qualitative factors, are used by the supervisory authorities to form a comprehensive assessment of the situation of individual banks. Since stress testing is just one of many factors considered by supervisors, the results for the banks tested must be analysed and interpreted in detail.

Stress tests are hypothetical exercises that are based on highly adverse macroeconomic scenarios and generally have the greatest impact when the economy is emerging from a protracted, severe recession, as in Italy. In designing the test, the EBA used a static balance-sheet approach, i.e., one that deliberately disregards the measures that banks would surely adopt over the 3-year period under consideration to mitigate the negative effects of the hypothetical shock. The authorities in charge of the exercise have clarified that, unlike in the past, any capital guidance derived from the stress test outcomes does not in itself constitute a binding capital requirement to be met by a certain date. In any case, the risk that the exercise could trigger procyclical effects in a fragile macroeconomic environment is to be avoided.

In response to the market uncertainty that followed the UK referendum, the Italian government notified the European Commission of its plan, which was approved, to guarantee new bank bond issues until the end of this year and only where strictly needed. As in the past, the banks that take advantage of the guarantee, which will be priced at market terms, must comply with the European state-aid rules.

European legislation also allows States to take precautionary measures, including recapitalization, based on stress test outcomes. The current situation, with its many risks to financial stability, requires a public backstop to be used only if needed, in full compliance with EU rules, bearing in mind the potential systemic effects of a future crisis for individual Member States and the euro area as a whole. The argument made by some at the European level that government intervention to support Italy's banking system should have been carried out in the past, as in other countries, does not take into account how the conditions of the national banking systems have evolved differently over time.

The countries that drew most heavily on public resources in recent years did so to address full-blown bank crises, associated with very large exposures to highly risky and opaque derivatives or with loans concentrated in notably overvalued real estate sectors. The situation in the Italian banking system, as also described in the IMF's Financial Sector Assessment Program (FSAP) of September 2013, was very different back then: Italian banks had not been affected by these issues. In Italy, the increase in NPLs has mainly stemmed from the weakness of the real economy, which has continued until recently, within a regulatory framework that in the meantime has been radically transformed. Today, in order to limit broad-based risks, responses may be needed that, looking forward, recognize the change in circumstances and exploit the room for manoeuvre allowed under the rules.

Any future measures to safeguard systemic stability must not serve as a pretext for delaying the important remedial action that is now demanded of intermediaries. Italian banks, especially those in difficulty today, face multiple and important challenges: from the more active management of NPLs in order to sell them or, if they remain on balance sheets, to improve recovery rates, to the achievement of higher levels of efficiency; from the exploitation of the opportunities offered by the technological-digital revolution to the necessary reorganization of branch networks and the launch of ordinary and, if necessary, extraordinary cost-cutting measures, including in relation to staff costs.

In the short term the recovery of profitability has been made harder by the repeated calls for a strengthening of capital, and more generally of loss-absorbing instruments, from the supervision and resolution authorities, also in order to comply with increasingly stringent rules. Regulatory changes will reduce banks' riskiness, endowing them with greater loss absorption capacity, increased liquidity and lower financial leverage. This will stabilize the entire system, making it more resilient to a crisis. But in the meantime, especially in some countries, the new rules risk denting banks' profitability - already beleaguered by the crisis years - even further, increasing the cost of funding and determining a reduction in the overall size of the banking system, with potentially adverse effects on the availability of funding for the real economy and on banks themselves. The right balance must accordingly be struck between micro and macroprudential considerations; international debate has recently paid closer attention to this question.

This same balance must be also sought in the context of the new European regulatory framework for the prevention and resolution of bank crises. The Single Resolution Board has begun work to define the minimum requirement for own funds and eligible liabilities (MREL) needed to absorb losses in the event of a resolution. It is important that the eligible instruments be explicitly subordinated to ensure a clear and rapid allocation of the losses; that an excessive burden not be placed on global systemically important banks, which are required to comply with the total loss absorbing capacity (TLAC) standards; and that the requirement be calibrated in such a way as to avoid procyclical effects. Its introduction must also take into account the time required for the market to absorb a large number of new debt issues.

* * *

The result of the UK referendum is affecting the euro area at a difficult time. The economic recovery has begun but is fragile. Unemployment remains high. Inflation is still very low, far from the levels consistent with the definition of price stability. There is the risk that the already widespread sense of dissatisfaction with the European project will grow. At this juncture it is even more necessary to ensure that further delays in introducing supranational intervention tools do not exacerbate the area's fragilities and make it more difficult to react appropriately to a shock or to prevent contagion.

We do not underestimate the signals of nervousness and anxiety from the financial markets, concerning Italian banks. Together with the other authorities we are acting with determination to promote effective market solutions. Faced with the risk that, in a highly uncertain environment, circumscribed problems could undermine confidence in the banking system, some form of public intervention cannot be ruled out. We trust in a collective commitment to overcome the current difficulties.