Across European banking systems, asset quality metrics are currently historically strong, particularly in relation to residential mortgage loans. Over the past decade, ultra-low borrowing costs and benign macroeconomic conditions--including declining unemployment rates--have fuelled steady growth in residential mortgage books across Europe. With the ongoing sharp slowdown in European economies and house purchase activity, S&P Global Ratings expects that banks will continue to see a rapid reduction in mortgage lending and related business activity this winter.

However, we do not see residential mortgages at the forefront of asset quality problems or credit costs for European banks over the next two years. A few factors underpin our view:

• European labor markets are robust, with unemployment rates expected to remain broadly stable under our economists' base case over the next two years. Rising unemployment has historically been a key driver of mortgage delinquencies in Europe.
• S&P Global Ratings economists don't expect a severe or abrupt correction in European housing markets (see European Housing Markets: Soft Landing Ahead, published July 13, 2022). While declining affordability--due to multi-year increases in prices and, more recently, rising interest rates--will certainly weigh on future prices, we see support from still-healthy household balance sheets following pandemic lockdowns, the very robust outlook for employment even as economies slow, and longer-term housing supply constraints.
• A slowdown in the growth rate of nominal prices--and even a correction in real prices--will not necessarily translate into higher delinquencies and credit costs for banks, as retail borrowers tend to prioritize mortgage payments in many European markets (for legal or economic reasons).
• We believe that European banks have a long history of dealing with mortgage risks and have strengthened their credit risk management frameworks over time thanks to lessons learnt from past crises, regulatory probing, and enhanced macroprudential policies in relation to mortgage lending.
• These factors could prompt banks to restructure mortgage lending terms (e.g., extending repayment periods or other forbearance measures), and we expect authorities would encourage or mandate such an approach to limit large-scale repossession of collateral and forced selling. While some forbearance measures could trigger a risk charge and therefore come at a cost for banks, we expect this to remain contained.

Our view of the potential resilience of European banks' mortgage portfolios also stems from our analysis of the 2021 European Banking Authority (EBA) stress test results. Among the 50 largest European banks that participated in the stress test--which envisaged a cumulative decline in nominal house prices of 16.1% over three years, way beyond our base-case and downside scenarios--the cumulative additional provisioning on mortgage loans was estimated at €30 billion, representing a manageable 6% of total annual operating income for three years. By itself, this level of additional provisioning would not represent a material credit event for European banks, but second-round effects on other credit portfolios (e.g., lending to small and midsize enterprises or unsecured lending) could be more pronounced.

That said, we still see significant latent risks from residential mortgage portfolios in many systems. This is because they represent a material balance sheet exposure for European banking systems, typically around 30%-50% (or more) of total customer loans (equivalent to around 250%-400% of Tier 1 capital). In most cases, European banks tend to keep mortgages on their book until repayment or maturity, given that European securitization markets remain less developed than in the U.S. As a result, rising credit risk in mortgage portfolios will lead to a commensurate rise in bank provisioning, and a direct hit to their earnings prospects.

From a rating perspective, we would likely reflect any major change to our assessment of the relative strengths or weakness in a given banking system's mortgage market in our Banking Industry Country Risk Assessment (BICRA) trends and/or scores. We have not made any changes so far, although we consider that some European housing markets have entered a period of correction. In the U.K., we have seen early signs of house price declines, at a time when the economy is entering a broad-based correction (see Robust Earnings And Balance Sheets Keep The U.K. Banking Industry In BICRA Group 3 Despite Deepening Economic Woes). In Sweden, we have recently indicated that real estate markets had now moved from an expansionary to a contractionary phase (see Swedish Banks Can Weather A Housing Market Correction; BICRA Group Remains '2'). That said, we do not consider these changes to be material enough to warrant a revised view of BICRA trends or scores in these two systems, given that we do not see the correction resulting in steep rises in credit losses.

Credit Quality Metrics For European Banks' Residential Mortgage Exposures Are At Historical Highs

Residential mortgages are a material exposure for most European banking systems. At an aggregate level, EU banks carried €4.1 trillion of residential mortgages as of end-March 2022, representing one third of customer loans. In fact, mortgage loan growth has outpaced other types of loans since 2020, up by 8% or €300 billion in total. Comparing across systems, mortgage loans represent between 20%-60% of total customer loans (see chart 1).

Chart 1

Unsurprisingly in a context of rising house prices, benign economic conditions, and elevated loan growth, mortgage loan quality metrics have improved over time across all European banking systems. Today, most European banks have managed their mortgage nonperforming loan (NPL) portfolios down to historically low levels of around 1%-2%. The main exceptions are Greece, Cyprus, and, to a lesser extent, Hungary, which still hold elevated--but largely reduced--shares of nonperforming mortgages. On average, European banks hold credit loss provisions covering around 25% of nonperforming mortgages, but with wide variations around this average.

Chart 2

Mortgage Losses Should Be Contained Across Most European Systems

Our economists consider that a sharp slowdown in European economies is imminent, mainly caused by an unprecedented deterioration in the terms of trade, which affects economic output with a typical time lag of two to three quarters (see Economic Outlook Eurozone Q4 2022: Crunch Time). For European banks, this will lead to reduced business activity and loan growth in the coming months, as well as a pick-up in credit costs, as a sharp contraction in consumer spending and falling investment will test the resilience of weaker corporate borrowers and households (see Rising Recession Risks Cloud Eurozone Banks’ Earnings Prospects).

We don't expect mortgage portfolios to be a major source of credit costs through the next phase of this credit cycle, however. This view holds even under our economists' downside case, which envisages an even-sharper, but still not prolonged, deterioration in economic output in major European countries (see Global Credit Conditions Downside Scenario: Recession Risks Deepen). This view holds for most European banking systems, although the underlying rationale for contained mortgage losses might differ.

In countries where overvaluation risks are the highest, banks benefit from relatively strong mitigants, which should limit mortgage losses (see chart 3). These include the strength of household finances, partly supported by public support schemes, as well as banks' prudent underwriting standards and proactive measures put in place by regulators.

On the other hand, in countries with traditionally weaker safeguards for banks--as evidenced by historically worse mortgage performance--we see limited signs of house prices overheating so far. Our base case for economic activity and employment in most of these countries remains broadly supportive of mortgage performance.

Chart 1

We assess house price overvaluation based on deviation from historical trends in terms of real housing prices and price-to-income ratios. We also take into account the modelled estimates of house price overvaluation disclosed by the European Systemic Risk Board. Based on these factors, we consider that six European countries face particularly high overvaluation risks: Austria, Czech Republic, Iceland, Luxembourg, Sweden, and the U.K.

We have therefore reviewed the strengths of three main mitigants that we believe would limit the impact of a potential residential housing correction on banks' balance sheets:

• The overall strength of the labor market and household finances, including social transfers and other forms of public support to households;
• The quality of banks' underwriting criteria for residential mortgages, including the share of floating versus fixed mortgage rates; and
• The regulatory measures in place, including macroprudential measures to limit mortgage origination and/or build countercyclical buffers.

Based on these metrics, the Greek and Cypriot banking systems stand out as particularly weak in Europe. The traditionally weaker performance of mortgage portfolios in these countries also supports our analysis. However, in both countries, the real estate markets are not showing signs of significant overheating at the moment.

Among countries with high or elevated overvaluation risks, we find that Iceland and Portugal have relatively weaker mitigants for banks. In Iceland, there is still a prevalence of inflation-linked mortgages--which can accelerate losses in a downturn coupled with high inflation. In Portugal, although overvaluation largely relates to properties bought by foreigners and not financed by local banks, we note that there is a still-prevalent share of floating mortgage rates (common among European banking systems, see below), meaning that interest rate increases will weaken borrowers' repayment capacity.

Factors Supporting European Banks' Mortgage Portfolios

We see unemployment, rather than falling house prices, as the key driver of mortgage delinquencies. Indeed, we expect that households facing financial difficulties will prioritize the most necessary payments (such as food, housing, and car payments). In most cases, defaulting on a mortgage would only happen in the case of more significant stress on household balance sheets, typically associated with job losses. Despite economic stagnation, our economists expect unemployment rates to remain broadly stable under our base case in 2023/2024 due to the historical strengths of European job markets (see charts 4 and 5).

Chart 4

Chart 5

We also believe that the relative financial strength of European households will support mortgage quality (see chart 6). In several countries where data on household debt show a relatively weaker position (such as certain Nordic countries and Switzerland), we believe that the overall wealth of the country provides a material buffer and that governments are able to intervene if needed--either via automatic stabilizers such as strong social safety nets or ad-hoc support programs.

Chart 6

We consider that European banks have been broadly prudent in their mortgage underwriting in recent years. Importantly, the share of residential mortgages granted on floating rates (or short-term fixed rates) has materially reduced over time and no longer represents a significant portion of newly originated mortgage loans, except in a few systems. Also, the use of mortgage interest rate hedges in some banking systems mitigates the risk from floating-rate mortgages. For example, while 95% of Finnish mortgages are floating rate, we understand that over one-third have seven-to-ten-year interest-rate caps or collars. In Italy and Spain, around 20% of new mortgage loans were issued with floating rates (or fixed rates up to one year), down from around 80% a decade ago. In other countries, such as France or Germany, the share of floating-rate mortgages is low to negligible. This shift in underwriting practice means that banks, rather than retail borrowers, bear and manage interest rate risks. Further, in those countries with floating-rate or short-term, fixed-rate mortgages--for example the U.K. and Sweden--banks typically assess repayment capacity based on a stressed level of interest rates, therefore building a buffer against a potential rise in rates.

We also believe that most European banks have strengthened their credit risk management practices, in part following regulatory investigations, but also based on lessons learnt from the COVID-19 pandemic. We expect banks, with encouragement or explicit direction from authorities, would proactively mitigate rising credit risk, for instance by proposing a concession on lending terms (e.g., extending repayment periods) for borrowers facing a cost of living squeeze. Depending on their terms, some of these forbearance measures could trigger risk charges for banks, but the overall credit cost would remain much more contained than in a case of borrower default, especially as collateral values decline.

Finally, national macroprudential authorities in Europe have also intervened to enforce limits on debt-to-income ratios, loan-to-value ratios, or both. While we don't think these measures are sufficient to keep mortgage growth and house prices in check, we believe they will help shore up bank balance sheets on the downside. Similarly, many authorities have implemented or announced additional capital requirements to ensure that banks build additional resources to face unexpected mortgage losses.

Mortgage Losses Would Clearly Rise Under A Stress Scenario, But Remain Broadly Manageable

To see how a deeper and more protracted recession would affect mortgage portfolios, we reviewed the results of the latest regulatory stress test conducted by the EBA in 2021. Under its adverse scenario---a sharp rise in unemployment and a meaningful decline in real housing prices--the test foresees:

• A three-year recession with a cumulative GDP decline of 3.6% in EU economies;
• A significant rise in unemployment--to 12.1% from 7.4% in 2020 (the base year for this test); and
• Residential real estate prices falling by a cumulative 16% over three years across the EU.

Among the 50 banks that participated in this stress test, the reported share of mortgage loans that would default under this adverse scenario more than triples over three years to reach around €200 billion, representing slightly under 5% of all residential mortgages. Provisions modelled by banks against risky (i.e., Stage 2) or nonperforming (Stage 3) mortgages rise by roughly €35 billion over three years (from €20 billion-€55 billion), meaning that coverage ratios would decline slightly to 20% from 25%. Spread out over three years, these additional credit losses would represent around 6% of EU banks' 2020 pre-tax operating income, an amount that banks would absorb via earnings rather than capital, mainly due to the revenue boost provided by rising interest rates.

Chart 7

Chart 8

Further, estimates of mortgage losses vary greatly across countries. This highlights once again that European banks benefit from a range of diverse mitigants and experience widely different mortgage losses, even when faced with a similar (hypothetical) shock. In most European systems (including major ones), mortgage losses would generally remain manageable under the EBA's adverse scenario, with nonperforming residential mortgages rising, but still below 5%. In two countries, mortgage NPL ratios could exceed 10%--Hungary (15.8%) and Ireland (14.6%).

Chart 9

Based on the EBA's latest stress test, we expect European mortgage portfolios to remain broadly resilient in the next phase of the cycle. However, a deeper or more prolonged downturn in house prices could hurt consumer confidence, and, with broader ripple effects from the economic situation, could lead to increasing delinquencies in other parts of banks' portfolios. In that sense, a correction in house prices may not result in a material increase in mortgage delinquencies but could still impede banks' asset quality from an overall weaker operating environment.

We Reflect Mortgage Risks In Our BICRA Scores And Trends, Which Are Broadly Stable

As part of our BICRA analysis, we assess the relative importance of mortgage risks for banks at a system level. In particular, the evolution of real house price growth is a key determinant in our assessment of economic imbalances, itself a key part of our analysis of economic risks in a banking system. A material change to our view of mortgage risks, or a fundamental shift in one of the main mitigants in a given sector, would most likely lead us to reconsider BICRA scores for that system.

For now, we believe that current BICRA levels and trends adequately reflect each European banking system's exposure to mortgage risks (see chart 10). In some countries, we already reflect elevated risks of mortgage performance via specific negative adjustments to the economic imbalance scores (e.g., in Iceland or Ireland). In Sweden, we have recently indicated that real estate markets have moved from an expansionary to a contractionary phase. We believe that the U.K. economy is entering a broad-based correction, with GDP contracting, inflation persisting, and house prices beginning to come under pressure after two consecutive months of nominal price falls. That said, we do not consider these changes to be material enough to warrant a revised view of economic risks in these two systems, given that we do not see the correction resulting in steep rises in credit losses.

Finally, the stable economic risk trend we see for most European banking systems reflects our view that those banking systems are well-positioned for the economic slowdown in our base case (see chart 10). In the case of Germany--which has the lowest possible economic risk score in our assessment--the negative economic risk trend is primarily due to the secondary effects of the Russia-Ukraine conflict, as well as continued global supply chain disruptions, rather than significant risks from the residential real estate market (see Geopolitical Risks Add Headwinds For German Banks, Despite Robust Capitalization).

Chart 10

Related Research

• Swedish Banks Can Weather A Housing Market Correction; BICRA Group Remains '2', Nov. 10, 2022
• Robust Earnings And Balance Sheets Keep The U.K. Banking Industry In BICRA Group 3 Despite Deepening Economic Woes, Oct. 25, 2022
• Global Credit Conditions Downside Scenario: Recession Risks Deepen, Oct. 12, 2022
• Rising Recession Risks Cloud Eurozone Banks’ Earnings Prospects, Sept. 30, 2022
• Economic Outlook Eurozone Q4 2022: Crunch Time, Sept. 26, 2022
• Geopolitical Risks Add Headwinds For German Banks, Despite Robust Capitalization, July 19, 2022
• European Housing Markets: Soft Landing Ahead, July 13, 2022