One of the central innovations of the Basel III bank capital regime is the introduction of a minimum required leverage ratio for regulated banks. The leverage ratio is defined as the ratio of Tier 1 capital to a measure of the total “amount at risk” known as the leverage exposure, 1 and the leverage ratio so defined is required to be at least 3%. This minimum leverage ratio requirement is intended to complement a revised earlier capital ratio, now known as the CET1 ratio, the ratio of core capital, Common Equity Tier 1 capital, to Risk Weighted Assets. The upshot is that Basel III specifies two different regulatory capital ratios, the new leverage ratio and the revised CET1 ratio, with two different measures of core capital as their numerators and two different “amount at risk” measures as their denominators.
It is odd, however, that Basel III has two different core capital measures rather than one. One must then ask how it can make sense to have two different measures of core capital, with one broader than the other. One must surely be better than the other. If the narrower one is best, then the broad one should be too broad because it includes softer capital items that the narrow one does not, and if the broader measure is best, then the narrower one is excessively conservative. The use of both measures is intellectually odd and creates scope for arbitrage, encouraging banks to game the difference between the two different measures.
The narrower measure is CET1 and the broader measure is Tier 1, where Tier 1 is defined as the sum of CET1 plus Additional Tier 1 (AT1) capital. Capital instruments are eligible to be classified as AT1 if they meet certain conditions, e.g., that they be issued and paidin, be perpetual and be subordinate to depositors, general creditors and subordinated debt. In practice, the AT1 instruments that matter most are Contingent Convertible bonds, known as CoCos, that convert to equity under certain conditions.
The question, then, and the focus of this article, is whether CoCos are suitable as core capital instruments. If they are, then CET1 would be excessively conservative and it would make sense to abandon it, but if they are not, then Tier 1 is excessively broad and it would make sense to abandon it instead.
This question is an important one because major claims have been made about the usefulness of CoCo bonds as a means of recapitalizing a bank in a solvency crisis. Investors in CoCo bonds take on the de facto role of providing capital “insurance”, providing for their investments to be converted into equity when it is most needed. This provision for contingent capital enables a bank to be recapitalized without it needing to raise additional equity on the market. If this contingent capital can be relied upon, then it would give banks a less expensive way of raising core capital when they need it most. But if it is not, the contingent capital provided by CoCos must be seen as inferior and CoCos should not be regarded as on a par with core capital. In this case, the Tier 1 capital measure that recognises AT1 instruments as core capital should be replaced and only CET1 should be recognised as “true” core capital.
This article is organized as follows. Section 1 explains the basics of CoCos: how they are structured, their purpose, how they work etc. Section 2 discusses some of the general issues and problems that arise with CoCos. Section 3, 4 and 5 examine specific CoCo topics in a little more detail: their triggers, their systemic stability issues and the lessons to learned from the experience of related capital instruments during the Global Financial Crisis (GFC). Section 6 examines the implications of declassifying CoCos as core capital instruments and Section 7 concludes.
- Kevin Dowd
Kevin Dowd (email@example.com) is professor of finance and economic at Durham University in the United Kingdom.