The European Central Bank (ECB) has just upped the ante on prudential provisioning. Earlier this year it introduced guidelines to exhort banks to reduce stocks of nonperforming exposures (NPEs), and now its new draft proposal aims to ensure that NPEs are sufficiently provisioned.
While we don't foresee material consequences for eurozone banks, we believe that if the guidance is adopted it could enhance the credibility of regulatory regimes and improve investor, and our, confidence in the strength of banks' balance sheets.
The new proposal would require banks to economically fully provision the unsecured portion of any new NPE two years after it becomes nonperforming. Likewise, collateralized exposures would have to be fully provisioned after seven years unless the collateral is realized in the meantime. The draft guidance appears consistent with EU authorities' efforts to strengthen frameworks for the management and reporting of NPEs across the region. Supporting the ECB's proposal are the European Commission's endeavours to introduce measures to help European banks work-out existing NPEs and prevent the future build-up of new ones: this also entails the possible introduction of minimum levels of provisioning, in terms of own capital deductions.
For eurozone banks, the near term consequences of the ECB's latest prudential provisioning expectations would be limited because they only apply to exposures that are newly classified as NPEs after Jan. 1, 2018. Moreover, new inflows of NPEs are rapidly declining across Europe amid more supportive economic conditions in the eurozone and we expect this trend to continue in 2018.
Over time, though, the guidance could have increasing repercussions for banks that are burdened by long loan work-outs and keep on their balance sheets large NPE stocks not covered by provisions. The effects would be felt more so by banks operating in Italy, Portugal, and Greece where banking systems have been characterized by less effective insolvency/foreclosure procedures and judicial systems, and given their higher reliance on future collateral values in provisioning. We expect, though, that these countries will continue to make progress in tackling NPE levels, noting the reforms and measures introduced over the last couple of years.
The ECB will ask banks to set aside more money if they keep NPEs for too long
We take a positive view of the ECB's focus on more stringent and clearer provisioning requirements. The guidance would not change accounting standards for NPE provisioning but the ECB is encouraging banks to close any gaps in their own provisioning plans by booking the maximum level of provisions possible under accounting standards. Where accounting provisions fall short of the ECB's guidance, banks will alternatively have to deduct any shortfall from their own funds when calculating regulatory common equity Tier 1 capital. If a bank sees good reason that it should provision less than the ECB's guidance, it will have to explain why it has deviated from the required backstops.
The ECB has also clarified its requirement that banks avoid cliff edge effects. Consequently, we expect banks will need to gradually build provisions for new NPEs in the coming years, until they reach full prudential provisioning. The new guidance, therefore, could affect regulatory capital ratios as soon as 2018 in our opinion. We expect the effects to be manageable in the near term, however, as we forecast new NPE inflows will remain low over the next two years amid current more benign economic conditions.
For most banks we foresee no impact on profitability. We expect that they will continue booking accounting provisions in line with their own assessments and existing accounting principles. Similarly, the ECB states that its proposed backstop guidance will not have any effect in the vast majority of cases.
The existing stock of NPEs remains the key asset quality concern for European banks. Press reports have mooted a discussion among European regulators about applying the new guidance to the more-than €900 billion of NPEs currently sitting on systemic European banks' balance sheets. While this would probably be welcomed by investors as a way to increase transparency and more-swiftly clean up balance sheets, we believe such a move would materially impair regulatory capital ratios. It would also challenge banks' ability to lend and thereby help sustain the EU's current economic recovery.
The ECB complements legislative efforts to strengthen legal and judicial systems for NPEs work-outs across Europe
Legal and judicial frameworks vary across Europe. In some jurisdictions the work-out of NPEs has been slower given certain characteristics. We refer particularly to where clear and timely debt enforcement processes are lacking, the duration and cost of foreclosures proceedings are onerous, and tax regimes are unfavorable. Despite the reforms introduced in recent years, NPEs' time to recovery in most southern European countries will remain substantial.
Ineffective recovery processes reduce banks' options to manage NPEs. First, uncertainties about time to recovery and recovery rates are what most depress the prices the market is willing to pay for NPEs. Second, inefficient judicial systems could encourage debtors to delay solutions. Recent data from the Bank of Italy, for instance, point to an almost linear correlation between recovery rates and time to recovery (see chart 1).
The new guidance might push banks to sell or manage NPEs more proactively
One way for banks to accelerate the resolution of NPEs, in light of the draft guidance, would be to sell off more. If banks are required to set aside more capital for NPEs not covered by provisions, they will have fewer incentives to keep long-dated NPEs on their balance sheets. Selling becomes a more preferable option: it's attractive from an economic/capital perspective because it frees up resources that can be used more productively.
We think that absent more demanding requests from the ECB and the local regulator, most banks will adopt a more gradual strategy of selling off selected portfolios and enhancing collateral. The reason is that for banks holding large stocks of NPE--say in Italy and Portugal--selling meaningful amounts of NPEs at current prices could force them to take capital strengthening measures to cover short falls. This is because market prices remain well below the long-term economic value that banks attribute to such assets.
In Italy, we expect NPE sales could reach €100 billion-€120 billion by 2019 or around one-third of the existing stock if all the planned transactions go through. The largest deals will come mostly from banks bailed-out by the Italian government (Monte dei Paschi di Siena, Veneto Banca, Banca Popolare di Vicenza). Unicredit's recent offloading of is likely to facilitate and accelerate negotiations between banks and creditors, it does not involve a transfer of these assets off banks' balance sheets. This means there will likely be no rapid step-change in the NPE stock. Also, if a distressed company under restructuring is liquidated, banks will still need to absorb the remaining losses after the recovery of collateral. Positively, the ongoing pick-up in real estate market prices in Portugal will likely help to reduce the gap between the transfer price and book price, reducing the potential capital hit for banks.
In countries where banks have already made progress in tackling asset quality issues--Slovenia, Spain, and Ireland for example--we expect the new ECB guidance to have a negligible impact on additional provisioning and on further accelerating the NPE sale process. New NPE inflows are low and NPE work-out efforts have been already successful.
We therefore anticipate portfolio sales to international counterparties will continue to facilitate the improvement in banks' asset quality.
In other countries, we do not anticipate banks will be incentivized to sell NPEs. In France, where the stock of NPEs remains contained albeit higher than in other large European economies, banks are generally inclined to keep them on balance sheet to facilitate judicial recovery or, more frequently, amicable out-of-court settlements that enable faster enforcement and consider social implications. The French loan sale market still needs further independent servicing capabilities; these are currently more limited than in other European markets. National insolvency and recovery proceedings in France are not an obstacle to the appropriate management of NPEs, in our view.
In Italy, banks might need to adjust their lending strategies
Overall, we think Italian banks are potentially most exposed among eurozone banks to the new proposal by the ECB because of the very long workout periods in the country. We see Italian banks becoming increasingly less keen to lend to domestic corporations, which are historically more sensitive to economic cycles and generate higher inflows of NPE in bad times. As such, they might be asked to hold more capital in the future.
The ECB has clarified that the trigger for full coverage--two years for unsecured and seven years for secured loans--will start as soon as a loan is reclassified as an NPE. The softest NPE definition usually starts with those unlikely to be paid but still paying, and past due by more than 90 days but still accruing. At a later stages, these NPE might move to the bad loan stock (sofferenze). Sofferenze are the weakest NPE in Italy: the already defaulted loans.
According to a recent Bank of Italy analysis, the banks took more than five years (from the time migrating to sofferenze status) to work out more than half of their sofferenze stock.
This means that these loans have likely been classified NPE for more than five years until being worked out. Therefore, we believe that for Italian banks the ECB proposal would result in a very large portion of future newly-generated NPEs falling within the time range of full coverage (table 1).
We have also tried to assess what the impact could be on other eurozone countries if they faced a stressed domestic environment. We have been doing this by looking at our assumptions on foreclosure periods and loss severity for archetypical pools of residential mortgage loans when rating residential mortgage-backed securities (see "Methodology And Assumptions: Assessing Pools Of European Residential Loans," published Aug 4, 2017). We took these assumptions and compared the timing and projected severity of losses with the pro rata progression of the ECB's proposed provisioning requirement over seven years (see chart 2). The loss assumptions reflect a substantial level of stress, equivalent to an 'A' stress scenario (see "Understanding Standard & Poor's Rating Definitions," published on June 3, 2009). We can conclude that for most markets--for which we have assumptions, the proposed prudential backstop would typically be less punitive in terms of timing and impact than our assumptions for such archetypical loans. Therefore, we believe the ECB's proposal, even if fully phased in, would not have a material impact on banks.
We note, however, that our analysis is based on residential mortgage loans only. For corporate loans with a potentially higher share of unsecured loans and therefore a shorter provisioning timeframe according to the ECB's proposal, the outcome could be different.
Against this backdrop, the Italian Banking Association has raised questions about the potential impact on banks' ability to support private sector activities if the ECB's guidance is introduced. We note, however, that Italian banks have tightened their underwriting standards since the economic recession, and we observe current keen competition for good quality customers. High quality borrowers' modest demand for credit explains the very low growth of loans to corporations in Italy, despite ample liquidity in the market. As of June this year, loans to corporations (net of disposals) in Italy increased by only 0.5% year-on-year after dropping by around 20% yearly since 2008.
The ECB's draft proposal has no immediate consequences for our ratings on European banks
Our ratings already factor most of the issues the ECB would like to tackle in this guidance. First, in our assessment of the risks faced by European banking sectors, we already capture structural weaknesses such as ineffective insolvency procedures and the length of time needed to recover NPEs. In our view, the large stock of NPEs will continue to weigh on banks in several countries despite recent progress. Second, compared to regulatory capital ratios, our risk-adjusted capital measures take into account the different economic risks to which banks in different countries are exposed and therefore tend to be materially lower (than regulatory ratios) in those countries that are currently suffering most from the high NPE burden.. Finally, we have made additional rating adjustments to outlier banks whose asset quality is considerably weaker than peers.
That said, we see the possibility that all the regulatory and accounting changes, including IFRS9, could provide new valuable information and lead to fresh evidence of weakness in some cases. We would incorporate any such evidence into our forecasts and ratings. We also expect that if the ECB guidance is adopted it could strengthen balance sheets and enhance the credibility of regulatory capital ratios. It could also help banks achieve more positive and less surprising outcomes in future stress tests.
Mirko Sanna - Primary Credit Analyst - Standard & Poor’s Financial Services
Francesca Sacchi - Primary Credit Analyst - Standard & Poor’s Financial Services
BLOG COMMENTS POWERED BY DISQUS