In this article:
Mopping up the entrails of a failed bank is a singularly complicated task. Now the Bank of England (BoE) wants the country’s top lenders to publish their ‘death’ plans, in a bid to increase the transparency around the process. Basically, the BoE wants to know exactly how such large and globally-interconnected institutions would go about winding up operations without generating a tsunami of market panic.
This is just another cog in the complex machinery of global regulations prompted by the financial crisis. In its aftermath, governments came under considerable pressure to ensure that taxpayers would never again have to cushion the fall of a collapsing bank.
Sharing the burden
One component of these regulations was the introduction of bail-in features, which allow, among other things, the statutory conversion of debt into equity. In a nutshell, GSIBs on the brink of failure can share the burden among their shareholders and bondholders, falling back on creditor-financed recapitalisation. Theoretically, this should stave off a disorderly insolvency that could, in turn, threaten the stability of the global financial system.
Bail-in conditions are activated when a bank reaches a ‘point of non-viability’ (PONV) and is therefore deemed ‘likely to fail’ by regulators. While discretionary, the PONV is partially tied to a bank breaching its minimum common equity tier 1 (CET1) with no recourse, or simply reaching a contractual trigger. A bank run on deposits, or by counterparties, could also signal a PONV.
Which assets can be ‘bailed-in’?
Of course, all of this has significant implications for those of us who invest in global bank bonds, particularly considering the range of assets captured by the rules. The bail-in-able stack of financial instruments includes common equity, preference shares, additional tier 1 (AT1) perpetual bonds, new-style tier 2 (T2) bonds, statutory subordinated senior debt, contractually subordinated senior debt, and structurally subordinated senior debt. In jurisdictions like the UK and Switzerland, grandfathered old style T1s, and operating-company grandfathered T2s are also bail-in-able.
Chart 1: How does bail-in work its way up the liability stacks?
We’ve been cognizant of the heightened risk around GSIB debt for several years now and have long since embedded this in our research process and quant tools, even before the first bail-in bonds were issued in 2014.
AT1 bail-in bonds are mostly classified as high yield (HY), so their presence in portfolios generally depends on their index inclusion and the perceived risk/reward for each particular issue, including extension risk, coupon cancellation risk, and bail-in risk, as well as associated quant signals. Demand for AT1 bail-in bonds can be volatile, which necessarily impacts decisions on allocation size; in some cases the answer could be none at all.
Bail-in senior and T2 bonds are prevalent in both investment grade (IG) and HY indices. Here we look at our forecasts, as well as relative value, picking those with the best risk/reward characteristics. Because demand for T2 and senior bail-in bonds is higher, due to broader index inclusion, ratings analysis and forecasting are also important in terms of pricing and performance.
Example: ratings analysis by bail-in stack
Pricing a bail-in bond
Arriving at an appropriate price for bail-in bonds involves a methodical and comprehensive analysis of the myriad factors that either contribute to or weigh on a bank’s current and future financial health. By investigating key variables such as profit and loss projections, leverage exposure, liquidity, and deposit stability, as well as accounting for foreseeable and one-off costs and charges, we can build a picture of an institution’s capital generation, capital buffers and funding resilience.
A bank showing signs of decreasing or thin capital buffers will screen poorly. Institutions with highly volatile returns, low profitability, poor asset quality, or exposure to conduct and litigation charges, are also more vulnerable to reaching a PONV and, potentially, resorting to burden sharing.
Source: Fidelity International estimates, Bloomberg, 13 June 2018. Reference to specific securities should not be construed as a recommendation to buy or sell these securities, but is historic data included for the purposes of illustration only. Investors should also note that the views expressed may no longer be current and may have already been acted upon by Fidelity International.
Our proprietary quant tools add a further layer of analysis, breaking down bail-in debt stacks into their sub-components, allowing us to compare the relative merits of each. Quant models also provide valuable insights into recovery scenarios, measuring coupon deferral and bail-in risk for each individual bond in a stressed downside event.
Meanwhile, our debt stack analysis helps us calculate what size net losses would spur a write-down of each component, from core equity through to senior debt. Broadly speaking, the larger the stacks, the better the recovery potential following a PONV event; particularly for senior securities.
Example: Quant stochastic model to measure LT coupon deferral and bail-in risk
Source: Fidelity International estimates, company reports, 13 June 2018. Reference to specific securities should not be construed as a recommendation to buy or sell these securities, but is historic data included for the purposes of illustration only. Investors should also note that the views expressed may no longer be current and may have already been acted upon by Fidelity International.
Markets take bail-in bonds in their stride
Bail-in-able bonds have opened up a new risk landscape for investors with exposure to large global banks. Certainly, T2s and bail-in senior debt spreads have widened relative to pre-crisis levels as it became clear that these were loss absorbing on a going concern, rather than gone concern, basis.
However, markets are relatively sanguine on the subject. While in some jurisdictions bail-in bonds are still being phased in (up to 2019), they appear to be well and truly priced in. AT1s issued in 2014 and 2015 carried large bail-in premiums with high single digit coupons; but these premiums have come down over time as more bail-in-able capital stacks have been issued.
As long as your analysis adequately reflects the evolving risk characteristics of such assets, we believe bail in bonds remain eminently investable instruments.