Following the failure of Saambou bank in February 2002, another seven South African banks failed within a month, including the fifth-largest, and a further five within a year. In total, twenty-two small and mid-sized banks deregistered over two years: half the total number of banks, and nearly 10 per cent of the deposit base. South Africa is one of the few jurisdictions that does not have a explicit deposit insurance scheme. While such a scheme may have prevented the first failure, I show that it would not have prevented contagion. The banks that failed were all well capitalised and solvent, but had relatively high levels of short-term funding from non-bank financial institutions. They would not have qualified for a retail deposit insurance scheme, and would still have experienced a run of non-bank funding. This highlights that deposit insurance is best seen as a tool that should be used for its stated purposes (protecting vulnerable depositors), and not as a general financial stability tool that can prevent contagion. Indeed, if agents expect that the authorities will use deposit insurance to ‘bail-out’ a bank, this would introduce moral hazard.
by 