Contingent convertibles are the idea of the moment in the seemingly never-ending debate on what to do about the banks. They are a rather neat solution to the problem of banks not having enough capital. Rather than forcing the banks to hold far more capital against potential losses – very simple, but also dangerous to the economy as the more capital they hold, the less the banks can lend – proponents of “CoCos” suggest banks issue bonds which convert into equity, boosting capital, when capital ratios fall below a pre-defined point.
Two weeks ago Lloyds Banking Group in the UK set out plans to convert £7.5bn of subordinated bonds into CoCos, which would turn into equity if the bank’s tier one capital drops below a certain trigger. This has the disadvantage of costing Lloyds a lot, as Neil Unmack points out: the coupon is up to 16 per cent. It is also, as PrefBlog says, “thoroughly insane” to link conversion to a measure which is going to change, but in so-far undefined ways. Gillian Tett, in the FT, fears their conversion could itself create panic.
But CoCos are a fabulous instrument for solving almost all the problems with the banking system at once. They just need to be used differently.
First, they give the banks the potential to receive new slugs of capital from the private sector, which should end the need for banks to go cap in hand to governments – as long as enough of their capital is convertible.
Second, they cost a lot. Yes, I know that sounds like a disadvantage. But one of the reasons they cost a lot is that they should be properly priced. The implicit guarantee of a government bailout if a bank hits a crisis is hard to put a value on (some 2002 research on Thai banks modelled it as a put option), but for bond holders is clearly high, and far higher than the payments made by banks in some countries, such as the US, for deposit insurance. With CoCos, that value is priced into the market: they should convert into equity long before any government bailout, so investors put zero, or very little, value on the implicit government guarantee. That has a positive knock-on effect: if bank capital is properly priced, the financial sector should shrink back to become a more reasonable proportion of the economy, helping rebalance the US and UK, in particular. (The cost should be far lower than the Lloyds bonds – the trigger is what matters, something I’ll come back to)
Third, CoCos should reduce the risk of a run on a bank. While they don’t have any direct benefits to liquidity, they should boost confidence by providing the (contingent) boost to capital and the assurance that the bank will not go bust if hit by big losses.
Finally, CoCos should give governments a way to convince investors (and taxpayers) that they will not rescue banks the next time they hit problems. It is hard to get this message across, because everyone knows that when faced by the collapse of the financial system, governments will step in. CoCos, though, provide a handy way to almost eliminate any need for governement involvement: just as long as there are enough of them.
The solution is not to have a small amount of extra contingent capital, as some are suggesting, and as Lloyds will have. Banks should be forced to turn a big chunk of existing bank debt into CoCos, with a conversion trigger based on a minimum level of capital – the worst case scenario – *or* on the bank failing. In effect, this would be a bank resolution regime which automatically converts debt to equity, avoiding the need for government rescues. Debt would cost a lot more, but bondholders would also have a far bigger incentive to keep an eye on bank management. Perhaps only 50 per cent of debt, or some other very large amount, rather than 100 per cent, would need to be convertible – but it would have to be far more than the normal belief of what might be needed in order to deal with the “black swan” events which lead to the need for government intervention.
Ordinary CoCos could still be produced, which would convert on various other triggers, such as a higher capital ratios, and so act to protect the more senior bondholders.
The disadvantage – apart from the cost to banks, which I see as an advantage – is that uninsured lenders to banks would be subordinated to customers of the banks, leaving those customers with less of an incentive to avoid weak banks. But given the Lehman and Bear Stearns failures happened when their customers jumped ship, it would be better for everyone if the customers were given priority.
Unfortunately, governments are unlikely to push this through, because the banks will be joined in their lobbying by all the investors who like being able to buy high-quality bank debt with an implicit government guarantee. But this solution would be far simpler and more effective than trying to break up the “too big to fail” banks, or trying to regulate them to ensure they don’t make mistakes.