A Solution to 'Too Big to Fail'?

The Financial Stability Board has today issued a consultation on a proposal for a common international standard on total loss-absorbing capacity (TLAC) for global systemic banks - the proposal is here and BBC news coverage is here. In plain English, G20 governments don't want their taxpayers to bear the losses associated with the failure of large banks in the future. For a long time, governments around the world have had a too-big-to-fail policy when it comes to systemically important banks.  Consequently, governments will save them at any cost when they get into trouble. Perversely, this too-big-to-fail policy actually makes banks more likely to fail since it encourages banks to take excessive risk in the knowledge that they will be bailed out.

Two key parts of the proposal are (a) to increase the capital requirements for the 28 or so globally-systemic banks and (b) to make resolution of failed banks easier by converting debt into equity and having living wills. I foresee at least three problems which make me sceptical. First, the solution has to be applied globally, but competition between countries will result in them lowering the standards applied to banks in their jurisdiction. This happened in the 1980s whenever Japanese banks were able to out-compete their Western rivals because of a laxer regulatory capital regime in Japan. Second, there is a time-inconsistency problem in that governments may say 'we will never bail out a large bank again', but when a bank gets into trouble, it is optimal for a government to renege on their prior commitment to not bail out banks. Third, these proposals misunderstand the role of capital in banking. As I argue in Banking in Crisis, capital exists to prevent banks taking too much risk ex ante rather than as something which absorbs losses ex post.