As the eurozone economy recovers, the hotly debated trillion-euro question is: When will the European Central Bank start to tighten its monetary policy?
Many market participants are impatiently waiting for the bank's June meeting in Tallinn for signs. Meanwhile, the market is hearing dissenting notes from some bank officials about the possible consequences of a drawdown of the ECB's €2.3 billion bond-buying program, not only for the region's economy but also its banking system.
S&P Global Ratings expects a gradual shift in the ECB's monetary policy that via higher interest rates should support a recovery in the profitability of most eurozone banks, but only modestly over the next two years. The central bank's accommodative monetary policy has been one of many factors that have weighed on European banks' revenue generation over the past five years. Now that the policy has passed through its zenith, the direction of travel seems clearer but we don't anticipate a big bounce in interest rates. The U.S. experience with the "taper tantrum" showed that sudden changes in monetary policy expectations can unleash market instability. A sharp reversal in monetary policy—although not our base scenario—or a hawkish market reaction to even a more gradual shift could destabilize some of the eurozone's more fragile banking systems at a time when the economic recovery in the region remains uneven.
The extent that an unwinding of the ECB's quantitative easing (QE) would increase banks' revenue generation and profitability will depend on factors such as:
- · The speed of the shift in monetary policy and its transmission to the economy;
- · The duration profile of a particular bank's assets and liabilities;
- · The secondary impact on a bank's credit demand and asset quality;
- · The impact on the pricing of non-lending assets, including sovereign bonds; and
- · The impact on the cost of funding.
We've already factored in a gradual tapering into our eurozone bank ratings as a baseline scenario. This assumption underlies our outlooks on the ratings, most of which are stable. Under this scenario, a continuation of the economic recovery and a gradual pickup in banks' ability to generate revenue would in turn support capital generation. A more abrupt shift in monetary policy, although unlikely at this stage, would be more detrimental to the currently more vulnerable banking systems. Such a shift could entail asset price devaluations (for example on sovereign bond holdings) or a broader deterioration in asset quality, for instance. The related impact would likely vary materially among countries and banks within a country.
In any case, we believe that a cyclical recovery alone is not enough to lift earnings of eurozone banks to a satisfactory level. To accomplish this, many banks still have to make material changes to their business models, including cutting costs further.
Profitability Might Not Rise Noticeably Before 2019
Under our baseline scenario, we expect the ECB will extend its QE program to 2018, but start reducing the volume of asset purchases to €40 billion a month in the second half of this year ("smaller for longer"). At the same time, we think the ECB will bring its negative deposit rate closer to zero in the latter part of 2017, but keep its main policy rate at the current level until at least 2019.
Such a scenario, including a steepening of the yield curve (see charts 1-3) should over time support revenue generation for most banks. This reflects the maturity transformation that underpins most banks' business models. However, we don't expect the impact to be immediate, and therefore, we expect the net interest margin of the eurozone banking system to remain under pressure into 2018 (see chart 4), with some improvements beyond that point. At the moment, only the long tail of the yield curve has been steepening. We therefore believe that banks with asset pricing of longer durations, such as French or Dutch banks, could benefit from the impact of QE in the next two to three years. In contrast, the impact should be less meaningful for systems skewed toward shorter durations, such as Italy's or Spain's.
Today's negative rate environment in the region has hurt the margins of particularly those banks with large deposit bases relative to their loan books. These markets, including Germany's, will likely benefit in the longer run from a higher interest rate environment. However, shifts in customer behavior may in many cases delay the benefit to banks' top lines. For instance, in Germany, retail investors might move funds from sight deposits back into longer-term products, which could delay the overall benefit of higher rates for banks by up to a couple of years. Also, banks have put into place medium-term hedges on their exposures to interest rate movements, which might delay the revenue pickup from non-interest-bearing resources (for example, equity and sight deposits). In other markets, mortgage customers who still have standard variable-rate mortgages could decide to move to fixed products that are less profitable for banks, limiting the revenue upside for banks.
Finally, higher long-term rates would also reduce banks' exposures to certain long-tail risks, such as defined benefit (DB) pension schemes. Until recently, market developments have continued to conspire against most European corporates with DB schemes, causing pension deficits to rise. The Netherlands, Germany, and the Republic of Ireland are examples of eurozone countries where private-sector DB schemes are more prevalent. Higher longer-term yields would significantly reduce DB deficits, primarily because the yield on long-term, high-quality corporate bonds is a key input in estimating the present value of DB schemes' liabilities.
A Tug Of War Between Steepening Yields And Decreasing Central Bank Support
To reduce the negative impact of its monetary policy on bank earnings, the ECB substantially increased its funding support to eurozone banks between June 2014 and March 2017. We expect that monetary tapering will likely be met with a reduction in funding support. The unwinding of this support in three to four years--absent new long-term support measures--will mitigate the net margin benefit to eurozone banks of a tighter monetary policy as they seek to replace this support via what is most likely to be more expensive refinancing in coming years.
The ECB's funding programs have been attractive for banks. On March 10, 2016, the ECB announced the launch of its second series of targeted long-term refinancing operations (TLTRO II), which took place every quarter between June 2016 and March 2017. The program primarily aimed to reduce bank funding costs and boost lending to corporates.
Broadly speaking, eurozone banks were able to take up four-year fixed-rate funding from the ECB, at very low or even negative rates, up to an amount equivalent to 30% of their stock of loans to nonfinancial corporations and households (excluding mortgages) as of Jan. 31, 2016, less any amount that they borrowed from the previous TLTRO (June 2014 to June 2016). The final TLTRO tender in March 2017 still raised €233 billion of subsidized four-year funding for eurozone banks, highlighting the attractive terms of this measure.
We expect that the increase in the cost of funding for most banks will be contained in the next two years. As of May 2017, the ECB still provided in excess of €750 billion of long-term funding to eurozone banks (see chart 5). However, the rise in funding costs should initially be contained as we believe that banks will use the current favorable market conditions to prefund some of the maturities. The gradual steepening in the yield curve relative to the long-term maturity profile of ECB funding they have received should give them some flexibility. What's more, a gradual repricing of assets will in our view match the somewhat higher cost of funding—though likely limiting some of the initial benefits to profitability.
We calculate that countries like Italy, Spain, and Portugal remain among the most dependent on this form of funding, though Greece is by far the main outlier. ECB funding represented about 6% of these countries' total banking balance sheets at end-March 2017, compared with a eurozone average of about 3%. We nevertheless believe that the ECB will continue to support liquidity at eurozone banks even after TLTRO funding winds down.
Decreasing central bank funding support will put an end to the carry-trade game played by a number of banks in the region, including those in southern Europe, over the past three years. This could over time provide an incentive for them to reduce their exposure to domestic government bonds, although liquidity requirements will likely mitigate this reduction.
The Dangers Of A Hawkish Market Reaction
We expect that even as it tapers, the ECB will continue its monetary balancing act. Lower rates and ECB bond purchase programs have benefitted not only banks, but also governments, keeping down financing costs for both. A hawkish market reaction to changes in the ECB's monetary policy could boost government bond yields, especially those of sovereigns with a weaker credit standing, as well as bank funding costs. The mark-to-market value of sovereign portfolios could decline, causing losses and capital adequacy challenges for banks.
In other words, there are still knots in the doom loop between sovereigns and banks in the eurozone. That's because, for example, most banks in the region hold high concentrations of national government debt in their liquidity portfolios. Eurozone sovereign debt represents on average about 5% of banks' balance sheets in the region and two-thirds of their capital and reserves (see chart 6). In countries such as Italy, Spain, and Belgium, this ratio was closer to or in excess of 90%. Any drop in the value of this portfolio could therefore have a meaningful impact on these banks. Higher market volatility could ensue, as illustrated by the "taper tantrum" in the U.S. in 2013, which could reduce valuations not only for government bonds, but also for most other securities as well, especially because many investors have gradually increased the duration of their fixed-income portfolios in the face of low interest rates. The lower valuations could in turn feed into material unrealized losses on available-for-sale portfolios of banks, potentially weakening their regulatory capital ratios.
A sharper increase in the cost of debt would likely also dampen credit demand, especially if wage growth remains subdued. Despite very low rates, private-sector loans have only started to recover modestly in the region as a whole since end-2014 (see chart 7). The indebtedness of the nonfinancial corporate sector remains on average high by both historical and international standards, constraining related credit demand. However, the high average masks diverging indebtedness and debt affordability levels within the region. A decrease in lending demand could offset banks' improving margins.
Will Asset Quality Withstand Higher Rates?
We believe that low rates have been instrumental in slowing the rise in nonperforming debt in certain countries and in allowing borrowers to adjust their debt profile. That said, Italy, Ireland, Spain, and Portugal, hit particularly hard by the financial crisis, still have a large stock of nonperforming assets (NPAs). In fact, their NPAs represented about half of the entire European stock at the end of 2016. Trends in the four countries differ, however.
For Irish and Spanish banks, improving economic conditions will continue to help reduce NPAs and, consequently, credit losses. Conversely, Italian and Portuguese banks' ability to work out their NPAs and quickly reduce credit losses continues to be undermined by a more-modest economic recovery and their relatively weaker solvency. The different paces of recovery in the region highlight the monetary conundrum for the ECB. A sharp increase in interest rates would likely hamper the timid progress being made in parts of the region to reduce NPAs.
Under our base case of a gradual increase in the policy rate starting only from 2019, we expect further improvements in the countries where asset quality deteriorated the worst after 2010. In the rest of the region, despite the continued recovery, we expect the cost of risk to gradually normalize from historically low levels (see chart 8).
In countries less affected by the financial crisis, a gradual steepening of the yield curve would benefit the stability of their banking systems over the longer run. It would also help normalize house price growth in cities where it's been rapid in the past few years, including in Germany and the Netherlands. Conversely, a protracted period of very low rates could fuel the formation of asset bubbles.
Certain banking segments may be more sensitive to even gradual monetary tapering, such as car finance and car leasing over the next two to three years. This follows a period of rapid growth in lending and car sales. Slower new car sales and an increased number of second-hand cars that will likely to come to the market, combined with loan refinancing at higher rates, could exacerbate potential pressure on asset quality in parts of the eurozone. Other sectors, including commercial real estate, have also traditionally been sensitive to changes in monetary policy. What should buffer some of the pressure on asset quality are stricter underwriting policies by banks in the aftermath of the financial crisis.
Debt Pricing Is Likely To Differentiate
We believe a steepening in the yield curve and curbs to purchases of bank debt by the ECB could reverse what has been a compression in funding costs. This has arisen as investors have searched for higher-yielding securities in the low interest rate environment. We've seen funding costs converge for different classes of bank securities, for example, senior and subordinated instruments, and also among banks themselves.
The timing of increased price differentiation is one key to bank profitability. We base our view on the substantial amount of subordinated debt—including new senior subordinated instruments—that eurozone banks will need to issue to meet recent regulations. According to the eurozone's Single Resolution Board, banks in the region may need to raise another €100 billion to €200 billion of subordinated instruments to comply with the new minimum requirement for own funds and eligible liabilities (MREL).
The ECB Is Testing The Effects Of Rate Shocks
The ECB is aware of the risks of the low-yield environment but also the risks of tightening. In its 2017 annual outlook and its May 2017 financial stability review, the ECB's Banking Supervision arm highlighted the low-yield environment as a key risk for eurozone banking systems. On Feb. 28, 2017, the eurozone's single supervisor unveiled details of two further regulatory initiatives for banks this year under its direct supervision, including 2017 Supervisory Review And Evaluation Process (SREP) stress testing. This seeks to understand the vulnerability of banks to interest rate risk in the banking book (IRRBB).
The stress test will simulate six hypothetical rate shocks to assess the resilience of economic value of equity (EVE) and net interest income (NII) of the 110 systemically important financial institutions (SIFIs) under the review. This approach follows the Basel Committee on Banking Supervision's IRRBB standards published in April 2016. The exercise follows a bottom-up approach in which banks under the review are requested to provide projections on their EVE and NII impacts, assuming the shocks would materialize in the banking book without adjusting any other parameters. ECB supervisors will then review the results with the banks under SREP discussions in July 2017.
The outcomes of the stress test will inform the ECB's assessment under its 2017 SREP and feed primarily into the Pillar 2 guidance (P2G) for individual banks. We understand that the resulting recalibration of P2G will be such that on average it will be neutral across all institutions. At the same time, banks with relatively high or low exposures could face surcharges on their P2G or benefit from discounts.
We believe that overall, more retail-focused and deposit-funded institutions could be more exposed to potential variations in capital requirements. This is because we consider that these banks' business models generally tend to rely more on earnings from mismatches in the maturity between assets and liabilities and are therefore more exposed to the interest rate shock assumptions. Furthermore, retail banking products, such as non-dated deposits, typically include more customer optionality, for which banks apply modeling assumptions in managing these risks. However, we expect the ECB to provide limited public disclosure of the outcome for specific banks. We expect that we will only be able to notice any add-on or relief to the P2G from the stress test, which the ECB will likely finalize in November 2017, if a bank materially adjusts its overall capital targets at that point.
Tapering Is No Substitute For Efficiency Initiatives
We believe a steepening of the yield curve alone won't deliver satisfactory earnings for eurozone banks, even if accompanied by a cyclical recovery. However, it should help, over time, given the maturity transformation function of most banks. Initially, we expect the positive impact of higher rates to be blurred by factors such as the reduction in central bank funding support, the normalization of cost of risk in more resilient countries, and interest rate hedging policies.
We therefore expect banks to continue their focus on efficiency gains. In some countries, further consolidation will likely play a role in supporting efficiency gains, but also in improving the banking system's competitive dynamics and the ability of banks to reprice lending. Absent such a trend, we expect that fragmented markets will be slower in reaping the revenue benefits of a shift in monetary policy. Similarly, reducing the stock of NPLs in the countries that were hit hardest by the financial crisis will become more critical than ever in anticipation of a normalization of monetary policy. Further changes to eurozone banks' business models therefore remain crucial for tapering to measure up to banks' earnings ambitions.