Taxpayer should not facilitate risky bank cocos

During the great financial crisis, banks appeared to be heavily undercapitalised. Banks had as little as 2 percent equity capital of risk-weighted assets (so-called tier 1 capital) and were allowed to count sub-ordinated debt as capital (so-called tier 2 capital). However, this subordinated debt did not absorb losses, when it was needed during the crisis.[1] To correct for that, the Basel Committee proposed higher and better quality capital for banks in the Basel 3 capital accord. The focus was on Core Equity Tier 1 (CET1) capital, which was increased from 2 to 4.5 percent of the risk-weighted capital ratio. So far, so good. But then the bank lobby started again. Under pressure from the US, the Basel Committee allowed cocos as additional tier 1 capital (as well as tier 2 capital). These cocos can contribute to 1.5 percent of the risk weighted capital ratio. Cocos are contingent convertible bonds that convert to equity if the regulatory capital ratio drops below a certain pre-determined threshold. More recently, the ECB (2016) has eased banks’ capital burden by replacing a portion of binding requirements with non-binding guidance.[2] This change makes it less likely that banks will face restrictions on dividends, bonuses and additional Tier 1 coupon payments. Why are cocos so popular with bankers? The tax-deductibility of interest has spurred the push towards the increasing use of debt as regulatory capital, in an attempt to have the best of both worlds - Telling the regulator that these ‘capital instruments’ can absorb losses, like equity, while at the same telling the taxman that these instruments are debt, so that interest payments can be deducted for corporate tax (Schoenmaker, 2015). The latter reduces the private costs of debt for banks. Cocos are thus an example of having your cake and eat it. More broadly, Allen et al. (2011) argue for equal treatment of equity and debt for corporate tax purposes. They note that it is not clear why in many countries debt interest is tax deductible at the corporate level but dividends are not. There does not seem to be any good public policy rationale for having this deductibility, which appears to have arisen as an historical accident. If tax deductibility is why there is a desire to use debt rather than equity, then the simple solution is to remove the tax deductibility. The same policy mistake again It looks like history is repeating itself. Before the crisis, subordinated debt was promoted by academics because of its disciplinary function (e.g. Flannery 2001). As banks with higher asset risk have to pay higher interest rates on subordinated debt, such debt can induce banks to lower asset risk in order to reduce interest payments. Next, indirect discipline may happen when regulators take prompt corrective actions against banks with high subordinated debt yields or banks unable to roll over subordinated debt. These corrective actions may not only prevent further losses of problem banks, but also stop bank managers from pursuing unsound risk. But it did not work as envisaged. First, subordinated debt yields were only partly rising, as investors (with hindsight rightly) expected to be bailed out. Next, subordinated debt (and bail-in debt) may work in the case of an idiosyncratic failure, but not during widespread banking crisis as authorities may not want to spread contagion by writing down subordinated / bail-in debt.[3] Several academics question the financial stability implications of bail-in debt and cocos. Avgouleas and Goodhart (2015) call for a closer examination of the bail-in process, if it is to become a successful substitute to the unpopular bailout approach. They argue that bail-in regimes will fail to eradicate the need for an injection of public funds where there is a threat of systemic collapse, because a number of banks have simultaneously entered into difficulties, or in the event of the failure of a large complex cross-border bank, except in those cases where failure was clearly idiosyncratic. Similarly, Chan and Van Wijnbergen (2015) show that while the coco conversion of the issuing bank may bring the bank back into compliance with capital requirements, it will nevertheless raise the probability of a bank run, because conversion is a negative signal to depositors about asset quality. Moreover, conversion imposes a negative externality on other banks in the system in the likely case of correlated asset returns, so bank runs elsewhere in the banking system become more probable too and systemic risk will actually go up after conversion. This is a form of information contagion. These predicted contagion effects have proved to be real. Deutsche was at the centre of the market volatility earlier this year in February when investors grew concerned about the potential for it to stop paying coupons on its additional tier 1 cocos because of a multibillion-euro loss in 2015 (FT, 2016). During February’s sell-off, the market for selling new cocos shut down entirely.