When the Spread Rises, the Flow of Credit Runs Dry

ACCORDING TO A STUDY ON THE 20102012 EURO CRISIS, BANKS REDUCED THE SUPPLY OF CREDIT IF THEY WERE VERY EXPOSED TO SOVEREIGN DEBT RISK AND RELIED ON INSTITUTIONAL INVESTORS FOR THEIR SHORT TERM FUNDING

by Filippo De Marco, Bocconi Department of Finance
Translated by Alex Foti

Over the last year, Italy has again been under threat of rising spreads on treasury bills, i.e difference between the yields paid by the Italian government and the German government to investors in their bonds. A rise of the spread signals how investors find Italian securities riskier and consequently value them less, asking for higher returns (in fact, yields and asset prices are negatively correlated). Many commentators argue that due to the spread, Italian banks, which have about €380 billion in government bond holdings, have increased rates on new mortgages and reduced the supply of credit to the real economy.

In this context it is useful to examine the experience of the 2010-2012 sovereign debt crisis, when the spread, not only in Italy. but also in Greece, Ireland, Spain and Portugal, reached very high levels, higher than current ones. Analyzing the data on sovereign debt exposure of the main European banks, I was able to provide econometric estimates of whether banks most exposed to spread variations actually reduced the supply of credit to businesses and increased rates on loans in the 2010-2011 period. The reduced availability of credit has had negative effects on private investment, especially by small and young firms, which tended to borrow more heavily from banks with higher exposure.

But what are the channels through which the spread damages banks and, ultimately, households and businesses? First, a decrease in the market value of the securities held in the portfolio can reduce the bank’s capital forcing it to cut credit to meet capital requirements. However, market losses are only potential, and depend on how the bank classifies government securities in its books. By simplifying a lot, these can be valuated at market prices (Marked-To-Market, MTM) or at historical prices (Held-To-Maturity, HTM). In the former case, but not in the latter, a rise in the spread immediately impacts the bank’s equity capital.

However, it is unlikely that the capital channel worked in this actual case. From the stress test data compiled by the European Banking Authority, we can see that in large part (about 50%) banks’ sovereign debt exposure was not MTM. Moreover, even among those MTMs, potential impairment losses were not accounted for in the calculation of prudential capital. However, the banks which were most exposed banks reduced their credit supply. How did it happen? It happened because the potential losses on sovereign debt increased the cost of short-term refinancing on the wholesale market funding channel. In fact, institutional investors who lend funds to banks react to potential losses, even if not yet realized. In other words: financial markets are not interested in the accounting definition of a bank’s sovereign debt exposure, but rather consider whether sovereign risk exists and if it can materialize in the future.

In support of this thesis I found that the American mutual funds which invested in unsecured securities issued by European banks (in particular commercial paper and certificates of deposits) were no longer willing to renew their loans to banks, especially if they were exposed to the sovereign spread risk. And it was the banks which relied on these markets for their short-term funding that cut the supply of credit the most. In conclusion: the spread does have effects on the real economy through the banking system.