Though their details deserve debate, the underlying logic of the US and European commercial bank stress tests is irrefutable. Commercial banking in normal times is profitable. In exigent circumstances additional capital is needed. The requisite amount of additional capital is a function of the specific exposures of each bank and the probability of a confluence of economic and market events.

This logic now needs to be applied to the world's central banks. Though the suggestion is purely hypothetical – no supra-national institution like the International Monetary Fund, World Bank or Bank for International Settlements has any such authority – such an exercise would reveal crucial fault lines in the global financial system and possibly help us remediate them. While the conduct of central banking remains a purely sovereign affair, there are growing and eerily similar tendencies between private and public-sector banks in their generation of risk and their mitigation activities. Central banks may not go “bust” per se, but they can lose money – lots of it. In a worst-case scenario, hundreds of billions in additional capital from taxpayers may be needed to restore their credibility.

Central banking is broadly profitable in normal times, largely due to seignorage income – the yield differential between notes and coins in circulation and the returns received from the fixed-income securities every central bank retains against these liabilities. But this has declined sharply as nominal government bond rates have fallen. Meanwhile, the stealth socialisation of risk via central bank balance sheets – through quantitative easing policies as well as foreign exchange intervention – has generated unprecedented and largely unhedgeable exposures. These accumulated risks now implicate some 20 per cent of global gross domestic product. Currency, credit and/or interest rate risks at some monetary authorities are now such that specific market movements would generate catastrophic losses.