The Bond That Bites Back – and Why Allocators Are Buying Again
Contingent convertible bonds – better known as CoCos – were designed to absorb losses so taxpayers wouldn’t have to. When a bank’s capital ratios fall below a regulatory threshold, these instruments either convert into equity or take a haircut on principal, depending on their structure. The logic is brutal by design: investors get higher yields in exchange for agreeing to stand in front of catastrophic bank losses. After the 2023 Credit Suisse collapse wiped out roughly 17 billion Swiss francs worth of AT1 CoCo bonds – while equity holders received some recovery – that trade looked like a very bad deal. For about eighteen months, the asset class sat mostly untouched by serious capital allocators.
That quiet period is ending.
A growing number of fixed-income desks are revisiting CoCo allocations, particularly the Additional Tier 1 (AT1) variety issued by European banks. The reasoning is not naive optimism about bank safety. It is the opposite: a cold-eyed recognition that the Credit Suisse write-down clarified the legal architecture of these instruments in ways that years of fine print never quite did, and that the yield premium now on offer reflects a risk that is real but potentially mispriced on the downside.
What Changed After Credit Suisse
The Swiss regulator’s decision to wipe out AT1 bondholders ahead of equity holders shocked the market precisely because it violated the conventional understanding of capital structure hierarchy. Equity is supposed to absorb losses first. When it didn’t in that case – due to specific provisions in Swiss law and the bond documentation – CoCo spreads blew out across the board, including for banks that had nothing structurally in common with Credit Suisse. That indiscriminate repricing is exactly what creates opportunity in fixed income, and allocators who stayed patient through the turbulence are now looking at spreads that remain elevated even as European bank fundamentals have stabilized considerably.
European regulators, particularly the European Banking Authority, moved quickly after Credit Suisse to reaffirm that within the EU framework, equity absorbs losses before AT1 instruments. That statement did not undo the Credit Suisse losses, but it did draw a legal line between Swiss regulatory discretion and the EU structure. For allocators focused on instruments issued under English or Dutch law by banks supervised by the ECB, that distinction matters operationally. The documentation risk is now more explicitly understood, not handwaved away.
What also changed is who is selling. In the months following the Credit Suisse event, forced redemptions and risk-off mandates pushed institutional holders to liquidate positions regardless of underlying credit quality. That kind of technical selling compresses prices in ways that have nothing to do with issuer fundamentals. The banks that saw their CoCo spreads widen most dramatically – large French, Spanish, and Nordic lenders with healthy capital ratios – were not in distress. They were collateral damage, and the securities they had issued were briefly available at prices that compensated for losses far more severe than their balance sheets suggested was plausible.
How Allocators Are Sizing the Position
The CoCo revival is not showing up as core holdings in balanced portfolios. It is appearing as a satellite allocation within broader credit sleeves, typically capped in the low single digits as a percentage of total fixed income exposure. The yield argument is direct: AT1 CoCos from investment-grade European banks have been trading at spreads wide enough to deliver total yields that meaningfully exceed comparable high-yield corporate bonds – without the same default risk profile, assuming the bank remains solvent and adequately capitalized. That is a different risk than credit default, and pricing the two as equivalent produces the mispricing opportunity.
Call risk is the other variable getting more attention now. CoCos are perpetual instruments but include call dates, and when banks declined to call them in 2023 – again partly Credit Suisse-driven contagion – the market treated non-calls as near-catastrophic signals. In practice, most European banks have since called their AT1s on schedule, and the non-call decisions in 2023 were concentrated in idiosyncratic situations rather than systematic policy shifts. Allocators building positions now are stress-testing call scenarios explicitly, modeling outcomes where extension risk materializes and ensuring the yield-to-perpetuity still justifies the entry price. For some issuers, it does clearly. For others, the math only works if the call happens.
Portfolio construction also involves thinking carefully about correlation. CoCos behave like high-yield credit in stress scenarios and like subordinated financial debt in normal markets – which means they tend to sell off sharply when either risk sentiment deteriorates or bank-specific news turns negative. That behavior makes them a poor hedge and a poor diversifier. They are a return-enhancement tool, and allocators treating them as anything else are misusing the instrument. The ones returning to the asset class appear to understand this distinction, sizing accordingly rather than building concentrated positions.
The Yield vs. Risk Arithmetic Right Now
The honest case for CoCos at this moment rests on a specific set of conditions that may not persist indefinitely. European bank capital ratios are, by most regulatory measures, at their strongest levels since the post-2008 reforms took full effect. The ECB’s supervisory regime has become significantly more prescriptive about minimum capital buffers, which raises the distance between current ratios and the trigger levels that would convert or write down AT1 instruments. A bank would need to lose a substantial portion of its capital base – not just have a bad quarter – before the CoCo mechanics activate. That buffer does not make the instruments risk-free, but it does affect the probability distribution of outcomes in ways that wide spreads sometimes fail to reflect.
There is also the income dimension that matters in a way it did not when rates were near zero. When a 10-year German Bund yields under 1%, a CoCo at 5% represented an enormous relative pick-up that attracted buyers who probably should not have owned the instrument. At current rate levels, the CoCo premium over safer alternatives is narrower in percentage terms but the absolute yield is higher, which means the investor base is more likely to consist of professionals who can actually model the downside rather than yield-starved allocators reaching into unfamiliar credit territory. A healthier investor base tends to produce less violent price dislocations when stress arrives – though it certainly does not eliminate them.
The Credit Suisse episode did permanently alter how AT1 documentation gets read, and allocators who ignored prospectus language before now treat it as primary due diligence. Jurisdictional differences between Swiss, UK, and EU-issued AT1s are no longer academic footnotes – they are material inputs into whether a position belongs in the portfolio at all. The banks still issuing CoCos know this, and new issuance documentation has become more investor-friendly in certain respects, clarifying trigger definitions and loss absorption mechanics. Whether that clarity holds under the pressure of an actual stress event is the question that a yield premium, however generous, can never fully answer.
