An intriguing idea from Alexander Campbell:
Why should we let banks get "too big to fail?" Why not simply impose exposure limits – not capital requirements, which can be gamed, but hard limits, backed by sanctions – set as a maximum possible share of the market? The government’s already in the business of deciding when banks (and other companies) are too big – by permitting or forbidding mergers that would give a single entity too much influence over its market, and by forcibly breaking up monopolies. All we need to do is focus in a little more, and we can avoid future equivalents of the discovery of Bear Stearns’ huge CDS counterparty position.
In practice this can’t really be done: there’s no way of getting there from here, since substantially all the major banks are at this point TBTF. And you can’t just break up a bank like you might break up a monopoly, along natural lines of business, since even if JP Morgan spun off its derivatives desk into an independent trading house, that independent trading house would still be TBTF.
All the same, I’d love to see an IMF paper on the financial systems of countries with many/few TBTF banks. If you were advising a developing country on bank regulation, would you advocate something along Cambell’s lines? Or do the advantages of having big banks outweight the disadvantages of having TBTF banks?