Results of the 2016 European stress test

The results of the stress test on the leading European banks, including the five largest Italian banks (UniCredit, Intesa Sanpaolo, Banca Monte dei Paschi di Siena, Banco Popolare and UBI Banca) were announced today. The test was coordinated by the European Banking Authority (EBA), working with the ECB and the national supervisory authorities. It is a rigorous exercise that tests the ability of Europe’s leading banks to withstand adverse economic and financial conditions with a low probability of occurrence.

The stress tests have been part of the banking supervisory authorities’ toolbox for a number of years. They can be used to assess the need for immediate capital strengthening. In the present case, however, the results will provide the supervisory authorities with information of use for day-to-day supervision.

The methodology of the stress test exercises can vary considerably from one country to another and over time. Thus, caution should be used when comparing exercises conducted by different authorities or in different years.

The present exercise is particularly rigorous, both in terms of its time span – three years: much longer than for similar exercises conducted in the US – and of some key methodological assumptions.

Features of the exercise

• Unlike the 2014 Comprehensive Assessment, this year’s stress test is not a pass/fail exercise as it does not stipulate a minimum threshold for capital that must be met by taking immediate strengthening measures. The results will be one of the elements used, though not automatically, to establish Pillar 2 capital at the end of the Supervisory Review and Evaluation Process (SREP). The decisions on the outcome of the SREP will be formally taken at the end of 2016 and will become effective at the beginning of 2017.
• The stress test assumes two scenarios for each country: a baseline scenario, which uses the European Commission’s autumn 2015 forecasts, and an adverse scenario. The simulation is conducted on the banks’ balance sheet data for the end of 2015.
• The adverse scenario for Italy assumes that real GDP will fall over the three years 2016-18 by almost 6 percentage points in relation to the baseline scenario forecasts. Thus, in 2018 output would be about 10 percentage points below the level registered at the onset of the financial crisis in 2007; such a drop would be without precedent since the last World War. The adverse scenario also assumes an increase of about 100 basis points in the yield on long-term Italian government securities in the three years, which would reduce their value by 12 per cent.

Table 1 – Change in the main Italian macroeconomic variables in the stress test
(per cent)

Source: ESRB/EBA, Adverse macro-financial scenario for the EBA 2016 EU-wide bank stress testing exercise, 29 January 2016. (1) Cumulative difference between adverse and baseline scenarios for GDP; average of differences between the levels of interest rates in the two scenarios.

• Coupled with the adverse macroeconomic scenario is a set of methodological assumptions, which are sustainable for the banks as a whole, but which can have particularly negative effects on banks still undergoing restructuring or already in a weak condition.
• As in past exercises the EBA has applied the static balance sheet principle. Accordingly, measures to shift portfolio composition towards less risky or more profitable products or market segments are inadmissible, and maturing funding that carries a very high cost cannot be replaced with less expensive new funding. In other words, in order to assess the resilience of the banks in the exercise according to a common analytical framework and based on comparable information, this approach deliberately overlooks any steps they would evidently take during the three years to counter the negative effects of the shocks.
• In this year’s exercise an idiosyncratic shock has been introduced as well, hypothesizing the bank’s immediate downgrading by two notches with irreversible effects over the whole of the three years considered. For banks that start with a low rating (B-, for example) the downgrading will significantly increase funding costs, by as much as 220 basis points for issues of senior debt. Moreover, it does not allow banks to pass on part of these costs to new lending.
• Briefly, the combination of static balance sheet and idiosyncratic shock heavily penalizes banks with a low rating by assuming new financing to be loss-making from the moment it is granted.
• The results for banks with a large stock of irrecoverable loans on their books for 2015 are further weakened by the fact that none of the interest maturing on those loans is entered in banks’ balance sheets. In fact, for the purpose of computing net interest income, the exercise considers that both exposures to insolvent parties and those for loans unlikely to be repaid, overdue or in breach of overdraft ceilings, do not generate interest, even though there is a non-negligible probability that borrowers in the last category will eventually resume regular repayments, as has indeed happened in previous years.
• Furthermore, the methodology ensures homogeneous comparisons between banks regardless of national differences in taxation by not allowing the recognition of tax assets created by temporary differences, such as from a write-down of available-for-sale financial assets (AFS). This rule magnifies the shock associated with a worsening of the sovereign risk on government securities in banks’ portfolios and penalizes banks holding such investments. As it is well known, the accounting criteria allow these differences; the body of prudential regulations, under ordinary circumstances, requires less harsh treatment.

Results for the Italian banks

• Despite the stringent nature of the exercise and the heavy tensions of recent years, four of the five Italian banks featured in the EBA sample show good resilience. For them (UniCredit, Intesa Sanpaolo, Banco Popolare and UBI Banca) the weighted impact on the CET1 ratio of the adverse scenario is 3.2 percentage points, against an EBA sample wide average of 3.8. With the inclusion of Monte dei Paschi, the weighted impact for the Italian banks comes of 4.1 points.
• Monte dei Paschi di Siena passes the test in the baseline scenario but obtains a negative result in the adverse scenario. The SSM has been monitoring the situation of Monte dei Paschi di Siena for some time. Since November 2013 the group has been following a restructuring plan, approved by the European Commission and still ongoing, which has led to significant progress in terms of more rational organization and cost cutting.
• About half of the overall reduction in Monte dei Paschi’s capital can be put down to a decrease in the interest margin; the other half is due to an increase in capital allowances and loan loss provisions and to write-downs on the AFS government securities portfolio. Two thirds of the P&L account loss comes from a decrease in net interest income. The idiosyncratic shock (equal to 220 basis points), which is proportionate to the bank’s initial rating (B-), is much greater than the shock assumed for higher rated banks (25 basis points for banks rated AAA), particularly since its effects continue for three consecutive years.
• The Board of Directors of Monte dei Paschi di Siena approved today a plan to sell its entire portfolio of bad loans and carry out a capital increase of up to €5 billion. This will make it possible to significantly increase provisioning for the remaining non-performing loans. As a result of this undertaking, Monte dei Paschi di Siena will have a stock on non-performing loans – but not bad loans – in line with the Italian banking system average. Its supervisory capital will remain at the current levels and profitability will improve both looking at funding and lending costs and at the return on assets and liquidity.

EBA - EU-wide stress testing 2016

EBA - Press release

ECB - Press release