Italy’s banks are being challenged by a rising tide of non-performing loans, and caught in the middle are retail investors who hold bonds sold by the country’s lenders. The political complications around the scale of small investors’ exposure are at the heart of the country’s ongoing banking woes.
Why does this matter?
State aid rules require losses for risky, or “subordinated”, bank bondholders before any government money is deployed. Additional new rules for failing banks require bondholders to take losses before state money is injected into the lenders.
These regulatory approaches are an alternative to vast government bailouts that occurred during recent crises, and that perversely shielded many bondholders. In short, under nearly all interpretations of the new various rules, some bondholders need to lose out to “unlock” the panacea of state aid.
However, many Italian bondholders are retail customers. In November, retail investors in four regional Italian banks took losses after subordinated debt was bailed in. This resulted in the suicide of a pensioner, led to a political backlash and ultimately transformed the debate around retail exposure.
How exposed are Italian retail investors?
An IMF report published this week says that retail investors own about a third of the €600bn of bank bonds in the country. They also own half of an estimated €60bn of subordinated bank bonds.
There are various kinds of bank bonds of different risks. Junior and subordinated bonds pay higher coupons but are exposed to losses before senior bonds, which offer buyers a lower interest rate.
How might retail investors lose money?
Under state aid rules, if there are no other possibilities for improving a bank’s capital position, “subordinated debt must be converted into equity, in principle before state aid is granted”. Italy wants to avoid this conversion process and add state funds without hitting any subordinated bondholders at all. These losses would apply only to the riskier bonds, which are only a proportion of total holdings.
How else might they lose money?
New rules introduced this year govern the “resolution” of a failing bank. When banks are placed into resolution, by their national central bank or the European Central Bank, senior bank bonds are then exposed to losses. So if Monte dei Paschi, for example, were placed in resolution, retail exposure would be higher than if only its subordinated bondholders were hit.
The new rules specifically require shareholders and creditors to take losses equivalent to 8 per cent of liabilities before state money can be used. The IMF calculates that for the 15 largest Italian banks, the 8 per cent requirement would hit retail subordinated debt investors in the majority of cases.
It estimates for about two-thirds of those banks, losses would be imposed on some senior debt holders in a resolution. It adds that this is just based on current liability structures, and that in a resolution case where capital has shrunk further, more senior debt could be hit.
The IMF sums up the conundrum: “Except in the case of very small banks, it is not clear to staff whether and how the current resolution framework will be implemented. That said, delaying resolution in cases of unviable banks can be costly.”
What does Italy want?
The Italian government wants to support its banks with taxpayer money while fully protecting retail bondholders — even those holding the most subordinated instruments. The November bail-in illustrated the severe loss of political capital imposed by retail losses. The country certainly wants to avoid a full-blown “resolution” of a bank.
Matteo Renzi, the country’s prime minister, has staked his career on a constitutional referendum in October. Further retail losses, even if only for the riskiest bonds and not necessarily systemic, could have a major impact on that vote.
One way in which retail investors could be spared entirely is by invoking an exception in state-aid rules, which means bail-in measures can be avoided if they stand to “endanger financial stability or lead to disproportionate results”. The question is whether losses for subordinated retail bondholders only would constitute such a threat.