Avoiding the next financial crisis

My new column for the FT argues that debating "too big to fail", the case for a new Glass-Steagall, and who does what in financial regulation is to miss the main thing: the need for capital requirements that vary with the credit cycle.

Remember that the US authorities, acting out of concern over moral hazard, let Lehman fail. In a way, they were right. It was not too big to fail: its collapse did not imperil the payments system and its counterparties did not fold. Yet praise for that principled decision was less than universal. Many argued, and continue to argue, that it was the worst mistake of the whole saga. The authorities are unlikely to forget this when another institution - which, regardless of its size, might be "too interconnected to fail" - looks ready to topple. And everybody knows it.

The precondition for big financial busts is always the same: unwarranted optimism. When everybody gets it into his head that inflation is tamed, interest rates will stay low, asset prices will keep rising and economic growth will never stop, overborrowing is sure to follow. In other words, moral hazard is only one factor reducing perceived risk. In a prolonged upswing, investors feel safe regardless - not because a bail-out will protect them from losses, but because they expect no losses.