When a financial crisis hits, regulators are often forced to bail out failing financial institutions, in particular large, ‘too big to fail’ ones. Putting a failed systemically important bank through resolution, imposing losses on depositors and other bank creditors, has generally been seen as too destabilising in the middle of a crisis. It is not surprising, therefore, that following the 2007-2009 Global Crisis, regulators were eager to encourage the creation of new instruments that facilitate automatic bail-ins – that is, an automatic deleveraging of distressed institutions.
Contingent convertible capital securities (CoCos), which are hybrid capital securities that absorb losses when the capital of the issuing bank falls below a certain level, have risen to centre stage as a potential bail-in device in future crises. CoCos can absorb losses either by having the principal written-off or by being converted into equity. The automatic loss absorption by CoCos when a pre-specified trigger has been breached was the main appeal of these securities according to their early proponents, in particular, Flannery (2005, 2016), Raviv (2004), Duffie (2009), McDonald (2013), Coffee (2011), Pennacchi et al. (2014), and the Squam Lake Working Group (2010). These advocates of CoCos disagreed on the optimal design of the instruments, especially on whether the trigger should be a stock price floor or a minimum equity-capital ratio, and their proposals were criticised by other commentators, most notably Admati et al. (2013) and Sundaresan and Wang (2015), who argued that CoCos were excessively complex and their conversion could have destabilising effects.
The introduction of the Basel III framework, which allows banks to meet part of their regulatory capital requirements with qualified CoCo instruments, created strong incentives for banks to explore CoCo issuances. As the number of jurisdictions implementing Basel III grew, banks responded by raising a substantial amount of capital in the form of CoCo issues. In a recent study (Avdjiev et al. 2017), we assemble a global data set of all CoCo issues by banks between 2009 and 2015 and use it to conduct the first comprehensive empirical analysis of this new market.
Between January 2009 and December 2015, banks around the world issued a total of $521 billion in CoCos, through 731 different issues. European issuers were early adopters and accounted for 39% of the CoCo market at the end of our sample. Banks from emerging market economies have experienced more rapid growth in issuance in recent years, and represented over 46% of the market by the end of 2015 (Figure 1). CoCos that have only a discretionary trigger account for 44% of total issuance volumes – the remainder have a mechanical trigger in addition to the discretionary one. The majority of CoCos with a mechanical trigger have capital trigger levels that do not exceed 5.125% of the bank’s risk-weighted assets, which is the minimum level required for a CoCo classified as a liability to qualify as Additional Tier 1 (AT1) capital under Basel III. The two loss-absorption mechanisms – principal write-down and equity conversion – are comparably represented in the sample. However, the issuance of principal write-down CoCos has picked up over time, most likely due to growing demand from fixed-income investors and because shareholder incentives to issue principal write-down CoCos are stronger (Figure 2).
Figure 1 CoCo issuance
Sources: Bloomberg, Dealogic, authors’ calculations.
Figure 2 The evolution of the CoCo market
Note: In some periods, there are minor differences between the quarterly issued amounts in the left- and right-hand panels due to incomplete information (on tier classification or loss absorption mechanism) for a small number of CoCos issued in the respective periods.
Sources: Bloomberg, Dealogic, authors’ calculations.
We show that larger banks and those with relatively strong balance sheets were among the first wave of CoCo issuers in advanced economies. Banks with impaired balance sheets that were in greater need of recapitalisation have been less likely to issue CoCos. Even though this result is surprising at first sight, it is consistent with the predictions of our theoretical model that issuance decisions are largely shareholder-driven. A CoCo issue by a bank with an impaired balance sheet mostly benefits the bank’s senior, unsecured debt holders. Consistent with this interpretation, we also find that the increased likelihood of banks with strong balance sheets to issue CoCos over those with weaker balance sheets is more pronounced for CoCos that absorb losses via a principal write-down rather than via a conversion into equity. Principal write-down CoCos are, in effect, junior to equity in distress states and, therefore, particularly attractive to shareholders.
We also analyse how CoCo issuance affects issuers’ balance sheets. We estimate the announcement effect of a CoCo issue on the issuer’s CDS spreads and equity prices in advanced economies, and explore cross-sectional variations by issuer characteristics and CoCo contract features. The overall impact of a CoCo issue on the issuer’s CDS spread is negative and statistically significant, which is an indication that CoCos are fulfilling their intended role of strengthening the issuer’s balance sheet.
When we separately examine CoCo issues by the loss absorption mechanism, we find that the impact on CDS spreads of issuing CoCos that convert into equity is much stronger than the impact of issuing principal write-down CoCos. This is consistent with the hypothesis that a CoCo that converts into equity disciplines risk-taking by shareholders because conversion may dilute existing equity holders’ claims. In contrast, risk-taking is rewarded at the conversion margin when CoCos absorb losses via a principal write-down.
We also find that only CoCos that have a mechanical trigger (in addition to a discretionary trigger) have a significant negative impact on the issuer’s CDS spread. In other words, the market does not display as much confidence that CoCos with only discretionary triggers reduce bank credit risk. The most likely explanation for this market reaction is that such CoCos are pure ‘gone concern’ instruments, with a lot of uncertainty surrounding the regulator’s conversion decision. CoCos that also have a mechanical trigger combine both gone concern and ‘going concern’ features, and hence offer a more reliable source of contingent capital. As expected, the impact of high-trigger CoCos on CDS spreads is stronger than that of low-trigger CoCos.
When we examine the announcement effects on bank equity prices, we find no significant impact of CoCo issuance on the issuer’s stock price over the full sample. This indeterminacy may reflect the opposing forces of ex post bail-in and ex ante risk-taking, as well as the potential dilution from equity-conversion CoCos. Interestingly, however, the one type of CoCo whose main design features are most beneficial to shareholders – i.e. a principal write-down CoCo with a high trigger – does have a positive and statistically significant impact on the issuer’s equity price.
CoCos are no longer a small niche market. Although they are deemed to be complex by many commentators, possibly too complex for retail investors, there appears to be a sufficiently large institutional investor clientele that stands ready to hold them. The change in the mix of CoCo designs is primarily driven by a combination of experimentation, issuer incentives, and investor demand. Now that the CoCo market is reaching maturity, our study offers the first evidence on which designs are desirable from a financial stability point of view, and where CoCo design can possibly be simplified with a view to standardising this market.
Admati, A, P DeMarzo, M Hellwig, and P Pfleiderer (2013), “Fallacies, irrelevant facts and myths in the discussion of capital regulation: Why bank equity is not expensive”, unpublished working paper.
Avdjiev, S, B Bogdanova, P Bolton, W Jiang and A Kartasheva (2017), “CoCo issuance and bank fragility”, BIS Working Papers no. 678.
Coffee Jr, J C (2011), “Systemic risk after Dodd-Frank: Contingent capital and the need for regulatory strategies beyond oversight”, Columbia Law Review 111, 795-1878.
Duffie, D (2009), “Contractual methods for out-of-court restructuring of systemically important financial institutions”, Submission requested by the US Treasury Working Group on Bank Capital, 9 November.
Flannery, M J (2005), “No pain, no gain? Effecting market discipline via reverse convertible debentures”, in H S Scott (ed.), Capital Adequacy Beyond Basel: Banking, Securities, and Insurance.
Flannery, M J (2016), “Stabilizing large financial institutions with contingent capital certificates”, Quarterly Journal of Finance 6, 1-26.
McDonald, R (2013), “Contingent capital with dual price trigger”, Journal of Financial Stability 9, 230-241.
Pennacchi, G, T Vermaelen, and C C P Wolff (2014), “Contingent Capital: The case of COERCs”, Journal of Financial and Quantitative Analysis 49, 541--574.
Raviv, A (2004), “Bank stability and market discipline: debt-for-equity swap versus subordinated notes”, unpublished working paper.
Squam Lake Working Group on Financial Regulation (2010), “An expedited resolution mechanisms for distressed financial firms: regulatory hybrid securities”, Squam Lake working group, Council on Foreign Relations, April.
Sundaresan, S, and Z Wang (2015), “On the design of contingent capital with a market trigger”, Journal of Finance 70, 881-920.