Microeconomics

Olivier Medjo
27 March 2010

Public guarantees in the wake of the global crisis have been wide-spread. This column presents recent research on the effects of a 2001 law to remove government guarantees for German banks. It finds that such guarantees were associated with significant moral hazards and removing them reduced the risk taking of banks, their average loan size and their overall lending volumes.

Do public guarantees influence bank risk taking? Public guarantees in the wake of the global financial crisis have been widespread. Many countries either nationalised banks (such as the US: Indy Mac, Fannie Mae, Freddy Mac; UK: RBS, HBOS, Lloyds; Germany: IKB, Hypo Real Estate; Belgium/Netherlands: Dexia, Fortis), or they provided blanked guarantees for the banking system (such as Germany and Italy) or both (Beck et al. 2010, Aït-Sahalia et al. 2009).

Despite this prevalence, there is scarce evidence on the likely effect of such interventions on bank risk taking.

This column reports recent research on the effects of public guarantees on bank risk taking and provides direct evidence that public guarantees are associated with substantial moral hazard effects.

Theory tells us public guarantees will have two opposing effects on bank risk taking. On the one hand, government guarantees may reduce market discipline because creditors anticipate their bank's bailout and therefore have lower incentives to monitor the bank's risk taking or to demand risk premia for higher observed risk taking (e.g. Flannery 1998, Gropp et al. 2006, Sironi 2003). This tends to increase the protected banks' risk taking. The effect is similar to the well-known moral hazard effect discussed in the deposit insurance literature (e.g. Merton 1977, Ruckes 2004). If depositors are protected by a guarantee, they will punish their bank less for risk taking, thus reducing market discipline.

On the other hand, government guarantees also affect banks' risk taking through their effect on banks' margins and charter values. Keeley (1990) was the first to argue that higher charter values decrease the incentives for risk taking, because the threat of losing future rents will act as a deterrent. Government bailout guarantees result in higher charter values for protected banks that benefit from lower refinancing costs. Hence, government guarantees may alternatively be viewed as an implicit subsidy to the banks that, through their future value, decrease bank risk taking.

Ultimately, the net effect of government bailout guarantees on the risk-taking of banks is ambiguous and depends on the relative importance of the two channels.

Whichever dominates is an empirical matter.

In 2001, government guarantees for savings banks in Germany were removed following a law suit. We use this natural experiment to examine the effect of government guarantees on bank risk taking, using a large data set of matched bank and borrower information.

We use a proprietary data set provided by the German Savings Banks Association for the years 1996 to 2006 that symmetrically spans the removal of government guarantees in 2001. The data set provides annual balance sheets and income statements of all commercial loan customers of all 457 German savings banks affiliated with the German Savings Banks Association. It includes data for over 87,000 customers. In total there are over 230,000 observations in the data set. The borrowers are largely small- and medium-sized enterprises with an average of €1.6 million in total assets. To control for savings bank characteristics, we also use annual balance sheets for the 457 savings banks. The savings bank data is also from the German Savings Banks Association.

As a measure for the credit risk at the borrower level we use Altman's Z-Score (Altman 1968) calibrated to the German banking market. In addition we analyse the effects of the removal of public guarantees on loan size, the charged interest rate spread, and the total loan volume.

The approach taken in the paper permits a unique identification of the effects of government guarantees on bank risk taking for a number of reasons.

• First, the removal of guarantees was exogenously imposed on the sample banks. The change in the safety net that we examine was unrelated to a financial incident, but rather based on a European court decision.
• Second, the banks in the sample are small and, therefore, unlikely to be "too big to fail". Hence, we can exclude the possibility that explicit government guarantees were simply replaced by implicit guarantees, which may have similar effects on bank risk taking and also be associated with moral hazard (Gropp et al. 2009).
• Third, the data permit a link between the balance sheet information of the banks and the balance sheet information of their commercial loan customers. Savings banks largely operate along traditional banking lines with little off-balance sheet operations.

Hence, we are able to measure their risk taking comprehensively by examining the Z-Score of their commercial loan customers.

Our main findings can be summarised as follows:

• The removal of government guarantees not only significantly decreased the risk taking of banks (the Z-Score of average borrowers increased by 7%), but we also show that after the removal of guarantees, banks reduced average loan size (by 13%) and overall lending volumes (by 22%). Riskier borrowers were either denied credit or were given a smaller loan. At the same time, banks increased interest rates for loans on the remaining borrowers (plus 57 basis points), despite their higher quality.
• We find that these effects tend to be significantly larger for banks, where it is likely that the ex ante value of guarantees was higher. First, the effects are larger for savings banks which were more risky before the removal of state guarantees. Ex ante riskier banks appear to have reduced their risk taking more after the removal of guarantees relative to ex ante safer banks. Second, savings banks which were associated with a more risky federal state bank also reacted more strongly to the removal of public guarantees. Our results show that the reduction in risk, the reduction in loan size, and the increase in interest rate spread were all significantly larger for savings banks that were associated with a riskier federal state bank.
• Finally, we show that the savings banks adjusted their risk taking both by dropping existing risky borrowers from their loan books (monitoring) and by tightening their lending standards for new borrowers (screening). The results suggest that some borrowers – generally the riskiest ones – lost access to credit due to the removal of guarantees

The results in this paper show that government guarantees are associated with strong moral hazard effects. The approach taken in the paper permits a unique identification of the effects of government guarantees on bank risk taking.