The future of AT1s:

The European Central Bank’s latest recommendations have rekindled debate over Additional Tier 1 (AT1) bonds. While authorities may question AT1s’ effectiveness in absorbing losses in times of market stress, legislative practicalities and the need to maintain AT1s’ market viability could limit the scope for change. With significant reforms unlikely, investors may instead focus on managing volatility, risk and opportunity in an environment of historically tight spreads in AT1 Contingent Convertible Bonds.

05 January 2026

ECB recommendation to enhance or phase-out AT1s, more conceptual than operational… at this stage…

Over the past few years, criticism of Additional Tier 1 Contingent Convertible¹ (“AT1 CoCos”) bonds from bank regulators, supervisors and standard setters has intensified. This has mainly taken the form of speeches and staff papers from officials at institutions. Beyond Australia’s decision to phase-out AT1 CoCos, finalised in 2025, concrete action taken so far has been limited. In Switzerland, recent proposals may materially increase UBS’ reliance on AT1s while a stricter approach to coupon payments and calls may be applied. This would be based simply on a stricter use of the Swiss regulator’s existing powers and would not require a wave of “new” AT1s.

The ECB’s recent set of recommendations to simplify the regulatory, supervisory and reporting framework² for European banks reignited the debate around AT1 CoCos’ role in banks’ capital structure. The European Central Bank (ECB) recommends either “enhancing” features of AT1 CoCos to ensure they are loss-absorbing in a going-concern scenario (ie before a bank is deemed failing or likely to fail) or removing AT1 instruments. The ECB itself emphasises that the latter is a non-starter Common Equity Tier 1 (CET1)³ capital requirements (replacing circa USD 230 billion* of AT1 capital with common equity) or a large decrease in overall capital requirements.

At this stage, we have confirmation that the ECB dislikes AT1s in their current format, although concrete regulatory changes have yet to materialise . Any change would require amending current EU regulation, done through a legislative process involving the European Commission, Parliament, and Council and typically involve consulting with a wide range of stakeholders. The ECB’s recommendations represent one of several inputs that will inform the legislative process. In 2022, the ECB officially voiced its support to strengthen the features of AT1s⁴, which did not translate into changes in regulation. Finally, any legislative process will likely be lengthy, hence any changes would be several years away at best.

Are AT1s already more loss-absorbing than it seems?

The proposal does not outline specific changes to the AT1 structure, and the ECB mentioned in the Q&A session that their current focus is on high-level recommendations. This has created material uncertainty and led to widespread speculation around potential changes to features.

The ECB’s recommendation is driven by their view that AT1s may not absorb losses effectively in a going concern scenario – before a bank is deemed failing or likely to fail.

Despite the criticism, AT1s are more loss-absorbing than it seems. Banks typically do not fail because of capital issues, but because of liquidity issues that indirectly deplete capital on a forward-looking basis (franchise erosion, higher funding costs, etc). A bank run occurs very rapidly, exacerbated by digital banking and social media, meaning that the line between viability and non-viability can be crossed in a matter of days. Bank resolutions, for example, Banco Popular in 2017, typically happen in a matter of days, leaving limited time for AT1s to play a role as loss-absorbing before authorities trigger the Point of Non-Viability (PONV)⁵. This ‘masks’ AT1s’ ability to absorb losses in a going concern as a bank run typically occurs before capital falls even remotely close to levels where coupons would be restricted, or bonds are written down. In cases like Banco Popular, AT1s and Tier 2 effectively absorbed losses and made the bank more attractive for an acquisition. More importantly, AT1s being written down has helped avoid a bail-out or imposing losses on more senior creditors – the former being a key objective of Basel III. Finally, there have been multiple cases of subordinated bonds taking a hit in a going concern – via distressed exchanges or tenders or coupons cancelled, for example.

Speculating on potential changes – a major overhaul seems unlikely

The crux with an aggressive change to AT1 features is keeping the asset class viable. Anything that materially subordinates AT1s to common equity would likely make the asset class non-investable and uneconomic for issuers. AT1s must remain more cost-effective than equity for banks to issue them, but also need to be less risky than equity for investors to buy them at that level.

Limited solutions to the AT1 trigger conundrum

AT1 principal loss-absorption (write-down or conversion into equity) currently occurs when: (1) the CET1 ratio falls below the trigger – typically 5.125% or 7%⁶ (2) at PONV if the competent authority (ECB and national supervisor) has assessed that the bank has failed or is deemed likely to fail.

Calibrating triggers to make AT1s more ‘going concern’ seems an impossible task. In a world where European banks’ CET1 capital ratios are circa 16% and minimum capital requirements around 10%⁷, triggers at 5.125% or 7% are irrelevant. Increasing these closer to the level of minimum requirements could be sensible, albeit in practice capital ratios are unlikely to drop below these levels before regulators trigger the PONV. However, if these are increased higher than requirements then this raises the issue of AT1s being written down or converted into equity in a scenario where the bank is not distressed. This would effectively subordinate AT1 to equity status, even in more ‘business as usual’ scenarios.

The idea of having liquidity triggers has also been floated, as capital tends to be backward-looking. While this may seem enticing, the choice of liquidity metrics, trigger level, etc. would be very challenging to calibrate. The use of “extraordinary” lines of liquidity under stress could be a trigger in itself, although at this stage PONV would likely already be reached, and hence powers to impose losses in this case already exist. In any case, a bank that cannot face outflows without extraordinary central bank or emergency government-backed liquidity lines is at least ‘likely to fail’.

As the ECB’s recommendation was formulated as a part of a wider “simplification” effort, perhaps removing mechanical triggers could be a solution, while using the flexibility offered by existing PONV triggers. Under the Basel framework, Additional Tier 1 capital does not require mechanical triggers, and, for example, neither US preferred stock, nor AT1s issued by Canadian banks have such triggers.

Flexibility on AT1 coupons is already extensive

Coupons on AT1 CoCos are fully discretionary and non-cumulative, meaning that issuers or regulators can cancel coupon payments, similar to dividend payments. Moreover, there is no hierarchy between dividends and AT1 coupons - AT1 coupons are not senior to dividend payments on common shares. Finally, coupons are subject to automatic restrictions if banks breach their MDA trigger (Maximum Distributable Amount), the point at which a bank’s CET1 capital drops below the combined buffer requirement.

Switzerland has proposed a profit test for AT1 coupons, where coupons would automatically be cancelled in case of a cumulative four-quarter net loss. While the idea is interesting, and a link between profitability and coupon payments makes AT1s more loss-absorbing, it raises several issues. Firstly, this may have undesirable effects, as earnings can be distorted by one-off items such as accounting changes, tax effects, goodwill impairments, M&A, etc. A mechanical link in this case would lead to coupons cancelled in cases where the need for more capital does not exist. Moreover, this could also lead to AT1 coupons being subordinated to dividends, if this applies only to AT1s.

The impact could be material. Looking at the AT1 market today (Bloomberg European Banks Tier 1 CoCo Index), 32 issuers representing more than 70%* of today’s market would have been subject to such a restriction at least once over the past decade. Chart 1 shows the percentage of the AT1 market (based on 14 December issuers’ weights in the AT1 market) that would have been impacted by restrictions on coupon payments, as high as 32% during Covid-19, for example. Assuming an average coupon rate of approximately 6.5%⁸, this implies a cumulative circa 7% (around 0.6% per annum) hit to AT1 total returns since 2014.

Raiffeisen Bank International (RBI) is a good cautionary tale; the Austria-based bank with subsidiaries across Central and Eastern Europe (CEE) recorded a net loss in the 12 months ended Q2 2025, which was driven by its Russian subsidiary. Had a profit test been implemented, AT1 coupons would have been cancelled, despite (1) capital ratios well above requirements even assuming a full write-down of its Russian unit (2) no liquidity issues and (3) the bank paid a dividend during that time. Any mechanical restriction on coupons in this case would have seemed unwarranted and unfair to AT1 holders.

Chart 1: Percentage of today’s AT1 issuers that would have been subject to automatic restrictions on coupon payments based on a rolling four-quarter cumulative loss.

Supervisors already have the necessary powers to shut off coupons, identically to dividends – this discretion can be used when warranted. Dividends are approved by regulators, based on capital plans submitted by banks, so why should the approach differ for AT1 coupons?

More of a non-event for existing AT1s than blue-sky scenario

The potential upside from a new wave of legacy AT1s seems limited. The most likely outcome is, in our view, either minor tweaks to AT1 features or the status quo. Even if we have minor tweaks, any implementation is several years away, with a likely grandfathering period. Any upside would come from lower extension risk, ie bonds trading at tighter spreads as existing AT1s would need to be called and replaced by new instruments, increasing the call likelihood. In today’s environment where spreads are close to all-time tights and around 100% of AT1s are priced to the next call – this limits the potential upside. Moreover, any grandfathering can be a double-edged sword, creating the risk of redemption of bonds at par (price of 100%). All AT1s have regulatory redemption features which provide issuers with the ability to redeem bonds before the first call at par in case these cease to count as Tier 1 capital. The average trading price of existing AT1s is around 103.5% today⁹, and over 80% of AT1s trade above par. Only 5% of AT1s trade below 95% (where there is large upside), while close to 40% of AT1s trade above 105%* (material downside). The devil is in the details, and some AT1s feature regulatory call clauses that can be exercised not only when bonds are disqualified, but when there is sufficient certainty of a future change. This means that some AT1s could be called early even if there is a long grandfathering period. To be clear, a very negative scenario is also not a base case.

Chart 2 – Split of the current AT1 universe by market price

Given the high uncertainty around potential changes and the lengthy timeline to implement these changes, we believe the impact for AT1 investors is limited, and the risk-return of any future changes seems balanced at this point. In our view, positioning in the asset class should remain focused on valuations and upside/downside risks – future changes in regulation are unlikely to mitigate potential price volatility in the near term. In the current context of spreads on AT1 CoCos close to their all-time tightest levels and close to 0% of AT1s priced to perpetuity, we favour a more defensive approach – focusing on senior and Tier 2 bonds versus AT1 CoCos.

Romain Miginiac is a Portfolio Manager and Head of Research at Atlanticomnium, and co-manages credit and sustainable bond strategies for GAM Investments.