Statement at the presentation of the Deutsche Bundesbank’s 2019 Financial Stability Review Speech delivered at the press conference to unveil the Financial Stability Review

Check against delivery.

This time last year, the economic outlook was slightly more positive than it is at present. Although economic warning signs were already apparent, interest rates were still expected to rise gradually. Unfortunately, this has not come to pass; downside risks to the economy have since emerged worldwide. Germany’s export-oriented industry has borne the brunt of this. Robust domestic demand – bolstered by favourable financing conditions and an upbeat labour market situation – has prevented an even greater economic downturn thus far.

Unlike in the previous year, market participants are now expecting risk-free interest rates to remain low in the years to come. A scenario in which interest rates are low for an extended period has therefore become more likely.

From the perspective of financial stability, this means that cyclical risks are continuing to accumulate within the German financial system. These risks are not directly visible as low interest rates are keeping valuations high in some cases, the markets are liquid, and asset prices and loans are still experiencing dynamic growth. At the same time, the vulnerabilities of the German financial system to unexpected macroeconomic developments are growing further.

We have described these vulnerabilities in past years:

  • After ten years of positive economic development, future credit risks are potentially being underestimated.
  • Meanwhile, the recoverability of loan collateral such as real estate is potentially being overestimated.
  • The situation is further compounded by interest rate risk: both an abrupt rise in interest rates and persistently low interest rates would put pressure on the financial system.

Two developments, in particular, could expose these vulnerabilities.

  • An unexpected economic downturn could hit Germany’s financial system hard. One potential trigger could be bad news concerning the global economy. Furthermore, external weaknesses may have a greater impact on domestic demand. Credit defaults would then increase and assets would lose value.
  • An abrupt rise in interest rates triggered, say, by the materialisation of downside risks and a sudden increase in risk premia in the markets, could also jeopardise financial stability.

A stable financial system should be able to cushion such shocks as much as possible rather than amplifying them. This year, the German Financial Stability Committee (FSC) thus recommended raising the countercyclical capital buffer to 0.25% on account of increased cyclical systemic risks.

Allow me to explain in more detail how we assess these risks and what action needs to be taken.

1 Underestimation of credit risk

Overall, the credit scores of German enterprises have been exceedingly good and insolvencies correspondingly low over the past few years. Hence, it follows that the number of non-performing loans has declined.

However, there are mounting signs that future credit risks are being underestimated:

  • Banks’ risk provisioning remains exceptionally low, reflecting the low number of insolvencies.
  • Economic downturns coupled with steeply increasing credit risks are likely to be underrepresented in many banks’ risk assessments. This is indicated, for instance, by the low risk weights of banks which use internal ratings-based models.
  • In addition to this, at around 5% over the past year, lending to the private sector reached a 15-year high. Large, systemically important financial institutions have recently expanded their lending in this area to a greater degree than small and medium-sized institutions.

Allocation risks may thus emerge for the entire loan portfolio that appear paradoxical at first glance: even if the creditworthiness of all enterprises has improved following a long upswing, making each individual loan appear sound, the structure of the loan portfolio may shift to enterprises that are riskier in relative terms.

Allocation risks, then, are not themselves a new risk, but rather a further indication that credit risks are being underestimated.

  • Creditworthiness varies between enterprises. Most enterprises have a fair credit score, although there are some which are rated particularly highly, and others which are weaker.
  • This distribution is broadly reflected in banks’ portfolios, too, although some enterprises also obtain funding via the capital market, whilst others draw on equity capital.
  • Now, what happens during a cyclical upswing? In such an event, all enterprises see an improvement in their creditworthiness. Strong enterprises increasingly turn to retained earnings or the capital market as sources of funding. Even relatively riskier enterprises are able to take out more loans from banks on the strength of an overall improvement in creditworthiness.

These are merely initial theoretical considerations which serve to illustrate the mechanism. But what concrete developments are we seeing in banks’ portfolios?

Here, enterprises’ creditworthiness has in fact seen an improvement. Compared with the beginning of the new millennium, enterprises’ interest coverage ratio – in other words, earnings relative to interest expenditure – has undergone continuous enhancement. Unsurprisingly, other indicators of creditworthiness present a very similar picture.

At the same time, comparatively risky enterprises have received more loans on account of their improved creditworthiness. The composition of loan portfolios has shifted. This is a statistical finding, first and foremost. It tells us nothing about the drivers behind this trend, such as higher risk-taking propensity among banks or a change in the demand for loans. Viewed in the context of banks’ creditworthiness assessments or from a supervisory perspective, this shift is not necessarily a problem.

However, banks are now increasingly financing the very enterprises that would be the first to encounter problems in the event of an unexpected economic deterioration.

What could happen in the event of an unexpected economic downturn?

  • Enterprises’ creditworthiness would deteriorate and more loan defaults would occur.
  • Loan losses and write-downs would increase more sharply than they would if credit risk were distributed more evenly.
  • Banks’ portfolios would then include a greater number of borrowers with lower credit ratings.

Such allocation risks are evidence of the fact that credit risk may potentially be underestimated following a long period of favourable economic developments. They have increased in recent years and are feeding into cyclical vulnerabilities throughout the whole banking system.

2 Overvaluation of assets and loan collateral

As market participants see it, interest rates are likely to remain low for longer than previously anticipated. Low interest rates bolster valuations in the global bond and equity markets. This is one reason why the weaker economic activity has barely made itself felt in the financial markets as yet.

At the same time, the banking system as a whole has become more vulnerable to an abrupt increase in risk premia. Should global political risks emerge or economic activity deteriorate unexpectedly, valuations would likely decline, meaning that losses would then have to be absorbed.

And in a low-interest-rate setting, asset prices react more strongly to minor changes in risk premia. Unexpected changes in market interest rates could thus be accompanied by a higher degree of volatility. This could potentially result in increased capital requirements for market risk.

Valuations in the real estate market are particularly significant for the German financial system. This is because more than half of all loans to households are loans for house purchase.[1] At 80%, real estate makes up the lion’s share of fixed assets in Germany.

Our assessment of the real estate market situation is similar to that of last year: cyclical risks, to which the real estate market is making a significant contribution, are continuing to mount. However, we see no need to activate borrower-based instruments intended to stabilise lending standards – in other words, a loan-to-value ratio and an amortisation requirement.

How have we reached this assessment?

House prices are an important indicator and have continued to grow sharply, rising by around 8% over the past year. Property prices in German cities are likely to stand around 15% to 30% above the value that would correspond to the fundamentals.[2]

At present, however, we do not see any clear indications of mounting speculation being driven by credit. Whether there is a self-reinforcing “narrative” – to use the words of Nobel Prize winner Robert Shiller – cannot be clearly assessed.[3] There are indications of this, however.

  • First of all, the competition in granting residential real estate loans is highly intense. With growth of 5%, loans for residential property recently grew at a slightly above-average rate. Segments of the market have shifted to the savings banks and credit cooperatives.
  • Prudential surveys amongst institutions show that there are certain signs of lending standards being relaxed – but only slightly and not across all institutions.
  • Households and banks expect residential property prices to rise further. It is difficult to say whether they are being overly optimistic here. At the very least, households are expecting less pronounced price rises in rural areas than in big cities and they expect smaller increases in prices per year over a period of five years than over one year.

All in all, it would therefore be premature to issue the all-clear for the real estate market. There is a need for action in two areas.

In a warning to Germany, the European Systemic Risk Board (ESRB) has just recently highlighted the need to close existing gaps in the data. We still know too little about one particular central indicator for estimating risk: the quality of lending standards.

Surveys of institutions, like those we have conducted once again over the past year, provide us with vital information. Ultimately, however, they are costly for all involved and do not offer a robust statistical footing for forward-looking policy. Over the past year, we have therefore continued to work on creating the preconditions for a legal basis regarding the collection of the necessary data. High-quality data are also urgently required in many other areas of public policy that concern the real estate market. However, it is policymakers who must ultimately reach a decision on the data collection recommended by the German FSC in 2015.

Furthermore, two additional income-based instruments should be created. Both instruments are oriented around indicators that banks use to assess creditworthiness in order to prevent borrowers from taking on excessive debt.[4] Both instruments likewise feature in the recommendations proposed by the ESRB and the German FSC for the German residential real estate market.

3 Interest rate risk

The German financial system remains vulnerable to an abrupt rise in interest rates. Many institutions have increased the interest rate fixation periods of their assets, not least to stabilise their interest income. For instance, the share of new loans for house purchase with an interest rate lock-in period of more than ten years has risen from 26% in 2010 to its current level of 50%.

Although this initially protects the borrower from rising interest rates, the risks do not disappear but are instead borne by the banks. A rise in interest rates would therefore impact many institutions at the same time. Funding costs would rise immediately, but interest income would only rise gradually. Savings banks and credit cooperatives would be affected to a considerably greater extent than larger banks.

4 Countercyclical capital buffer

These cyclical risks can emerge together and reinforce each other. An unexpected economic downturn could be amplified if banks restrict their lending. In order to prevent dangers to financial stability, banks must maintain capital buffers designed to absorb losses.

Macroprudential capital requirements serve precisely this buffer function. Under certain circumstances, banks can undershoot macroprudential buffers without measures such as the withdrawal of their banking licence being immediately imposed. Instead, bonus payments and profit distribution would be restricted. Macroprudential policy therefore complements supervisory measures that aim to ensure robust capitalisation and good risk management amongst banks.

The countercyclical capital buffer increases the resilience of the banking system and stabilises lending during periods of stress. Should cyclical systemic risks materialise, the buffer can be lowered straight away

In May 2019, the German FSC recommended that the Federal Financial Supervisory Authority (BaFin) activate the countercyclical capital buffer. Germany is thereby following in the footsteps of other European countries that have activated or increased the countercyclical capital buffer. BaFin followed this recommendation and raised the buffer to 0.25% of risk-weighted domestic exposures in the third quarter of 2019. Banks have 12 months to build up the buffer, and, for the most part, they may use their surplus capital to meet the requirement.

The recommendation makes allowances for uncertainty about economic developments. Future adjustments to the countercyclical capital buffer will depend on whether and to what extent cyclical risks continue to build up in the financial system. The financial cycle, and not the economic cycle, is the relevant factor here.

5 Need for action to ensure a stable financial system

Digitalisation, climate change, demographic change, and globalisation are megatrends that are impacting all areas of the economy and society. The financial sector mirrors these developments and, at the same time, is a key catalyst for the necessary structural change.

For the first time, this year’s Financial Stability Review features a chapter on the impact of climate-related risks on the financial system. In this context, our goal is not to evaluate climate-policy decisions. Instead, we aim to analyse the impact of climate change as well as the political, social and economic responses. To this end, more than 45 central banks and supervisors worldwide have joined forces in the Network for Greening the Financial System (NGFS), pooling their methodological capabilities and building up a robust information base. At the international level, the Financial Stability Board’s (FSB) Task Force on Climate-Related Financial Disclosures undertook important preliminary work for CO2 risks to be better disclosed in enterprises’ balance sheets.[5]

In order to tackle climate change effectively, it is necessary to implement a policy regime that adjusts the relative prices between carbon-intensive and low-carbon economic activities. This, in turn, necessitates significant structural change. A functioning financial system is key to ensuring that structural change can succeed without hampering innovation or jeopardising stability. 

However, structural change has also not stopped short of the financial sector itself. Banks – just like firms in other economic sectors – must therefore be allowed to withdraw from the market or to shrink if their business strategies are no longer sustainable. An effective resolution and restructuring regime can help market mechanisms to function properly. With the banking package recently passed by the EU, key regulations on creditors participating in losses – known as a bail-in – have been brought into line with international standards.[6] Also at the international level, the FSB is currently evaluating the impact of too-big-to-fail reforms.[7] A corresponding consultative report will be published in June 2020.


  1. This figure refers to lending to domestic households and enterprises.
  2. The Bundesbank estimates that house prices in German towns and cities were overvalued by between 15% and 30% in 2018. See Deutsche Bundesbank (2019), Monthly Report, February 2019. The fundamentals include per capita real estate holdings, per capita income, population density, and the age structure of the population.
  3. See Robert J. Shiller (2019), Narrative Economics: How Stories Go Viral and Drive Major Economic Events. Princeton University Press. Oxford and Princeton.
  4. Relevant indicators include, amongst others, aggregate debt as a percentage of household income as well as the percentage of household income needed to make interest and redemption payments.
  5. See Task Force on Climate-Related Financial Disclosures (2017), Final Report, Financial Stability Board.
  6. In this context, the minimum requirement for own funds and eligible liabilities (MREL) was revised.