VoxEU Column EU policies Financial Markets International Finance

Is the recent bank stress really driven by the sovereign debt crisis?

Stress in the interbank market has increased dramatically since July 2011, and bank stock market valuations have fallen by 22% on average for 60 of the most important banks subject to stress tests. This column argues that bank stock valuation has been affected by the banks’ exposure to Greek debt and that Greek banks were particularly affected. Holdings of debt of the other four periphery countries does not, however, appear to be a strong determinant of stock price movements.

Stress in the interbank market has increased significantly since July (Figure 1). There is now a significant debate on why this is the case and what would be the best way to address it (Financial Times 2011).1 Many have argued that the sovereign debt crisis isthe most important driver of banking stress in the Eurozone. If that view is correct, then the right approach to solving Europe’s banking problem is to solve the sovereign debt crisis. Recapitalising banks instead would be far too costly, in particular if one wanted to cater for a haircut in Italy.

Figure 1. Euribor-Eonia swap spread: Unsecured overnight lending becomes more expensive

Source: Calculation with data from Datastream.

In this column, I investigate to what extent the strong decline in bank market capitalisation observed since July can be explained by the exposure to sovereign debt of the five different periphery countries. Stock prices of the 60 most important banks tested in the European Banking Authority stress tests have fallen on average by 22%, with some banks losing as much as 80% of their market value. I relate the stock market fall of these 60 banks to the exposure of the specific bank to sovereign debt holdings of Greece, Portugal, Ireland, Spain and Italy.

In a first attempt, I plot the average weekly change in the stock market index against the total exposure of sovereign holdings of all five periphery countries as a percentage of the banks’ tier one capital. The graph (Figure 2) reveals that no strong correlation can be found. The graph also shows that many banks hold sovereign debt of the five periphery countries in excess of 100% of their Tier 1 capital.

Figure 2. Change in stock market index to total exposure to 5 periphery countries as a percentage of Tier 1 capital

Source: Calculation with data from EBA, July 2011, and Datastream.

As this figure does not account for differences in the holding of the five different countries, I plot the exposure to Greece and to Italy separately. Figure 3 suggests that exposure to Greece is a significant determinant of bank market capitalisation. However, the negative correlation is driven by a small number of banks that hold very large amounts of Greek debt. These are the Greek banks themselves, and indeed the correlation is significantly weakened when one excludes the Greek banks from the sample (Figure 4).

Figure 3. Change in stock market index to exposure to Greek sovereign debt as a percentage of Tier 1 capital

Source: Calculation with data from EBA, July 2011, and Datastream.

Figure 4.Change in stock market index to exposure to Greek sovereign debt as a percentage of Tier 1 capital, Greek banks excluded

Source: Calculation with data from EBA, July 2011, and Datastream.

In turn, exposure to Italian banks does not appear to be a significant determinant of bank stocks’ valuation (Figure 5).

Figure 5. Change in stock market index to exposure to Italian sovereign debt as a percentage of Tier 1 capital

Source: Calculation with data from EBA, July 2011, and Datastream.

These partial correlations suggest that the stock market valuation of banks has basically not been affected by their exposure to Italian sovereign debt, while the exposure to Greek sovereign bonds has played a role, even though this role is mostly an issue for the Greek banks themselves. To shed further light on the issue and check the robustness of this result, I perform a regression analysis. This allows simultaneous testing of the exposure to all five periphery countries, while at the same time controlling for the tier 1 capital ratio and the size of the bank as well as the location of the bank. The results of the regression analysis confirm the evidence presented and are detailed in my recent Bruegel policy contribution.

Conclusions

The results entail important conclusions for the current debate about bank recapitalisation. First, the recent massive decline in market capitalisation of banks does not appear to be driven primarily by the banks’ holdings of sovereign bonds. It would therefore be wrong to see the bank recapitalisation as a way to prepare banks for losses on their sovereign debt only. On the contrary, investors appear to accept the basic story that only Greek bonds will face a haircut while all other bonds will be serviced. It is of great importance to avoid a situation where investors believe that EZ leaders are preparing for losses on other EZ sovereigns. Second, there is nevertheless a significant loss in market trust in banks in the Eurozone. Sizeable bank recapitalisation could be one way of restoring trust in the EZ banks. However, it also appears plausible that this will not be enough. A number of factors could explain the recent mistrust of EZ banks. Markets might now perceive the general business model as no longer viable given a re-assessment of the risks. Also, the regulatory framework has so far rested on the assumption that sovereign debt is risk-free. A reassessment of this may lead to a complete overhaul of the balance-sheet composition of banks. Finally, there now appears to be a general crisis of trust on the part of some international investors in the viability of the Eurozone. EZ leaders should therefore map out a credible strategy for the institutional changes that are needed in order to render the Eurozone set-up viable. This will be the best way of solving the EZ crisis.

Author's note: Excellent research assistance by Chiara Angeloni is gratefully acknowledged.

References

European Banking Authority (2011), “2011 EU-Wide Stress Test, Aggregate Report”.
European Banking Authority (2011), “2011 EU-Wide Stress Test: Methodological Note- Additional guidance”.
Financial Times (2011), “America’s blueprint for saving Europe’s banks”, The A-list, 11 October.
Wolff, Guntram B. (2011), “Is recent bank stress really driven by the sovereign debt crisis?”, Bruegel Policy Contribution.


1 Roger Altman in the Financial Times of 11 October 2011 has called for a strong and ambitious bank recapitalisation strategy. Others, including Daniel Gros in his response to Altman, have argued that putting capital into banks would be wasted money as the current banking stress essentially results from stress related to sovereign bonds.
 

420 Reads