When Ownership Exists Only on Slides

https://global.morningstar.com/en-gb/stocks/credit-suisse-s-demise-a-timeline-of-scandal-and-failures

This is not a story about a bank that had no governance. It is a story about a bank that had nothing but.

There is a particular kind of organisational comfort that comes from a well-constructed accountability framework. The RACI is complete. The process owner is named. The governance structure is documented, reviewed, and signed off by people senior enough to make it feel serious. The slides are clean, the boxes are filled, and if you were to present the model to an external audience, it would look genuinely like an organisation that has its house in order.

Credit Suisse looked exactly like that organisation. For years.

Between 2018 and 2022, FINMA - the Swiss financial regulator - conducted 43 preliminary investigations and 108 on-site reviews of the bank. The governance infrastructure was not absent. The oversight mechanisms were not invisible. The accountability structures existed, were documented, and were regularly reviewed by the appropriate people in the appropriate forums.

And then something went wrong. Then something else went wrong. Then something else again.

In 2021 alone, Greensill Capital and Archegos Capital Management collapsed, exposing the bank to billions in losses. The bank was fined $475 million for its role in a Mozambique corruption scandal. Its long-standing chair stepped down. The following year brought further scandal, a criminal conviction, and a forced CEO departure. By 2023, a 167-year-old institution had been sold to a competitor for a fraction of its peak value.

The question worth asking is not what went wrong. The question is how an organisation with that volume of governance activity produced that volume of failure - and what it tells us about the difference between accountability that exists on slides and accountability that exists in practice.

The Architecture of Plausible Accountability

Ownership, as most organisations currently practise it, is a documentation exercise.

Someone is named. A role is defined. Responsibilities are listed with enough specificity to survive a governance review and enough ambiguity to survive contact with reality. The result is a model that is coherent on paper and negotiable in practice - which is, for most purposes, exactly what it was designed to be.

At Credit Suisse, the accountability architecture was not unsophisticated. Following losses in a relationship with a hedge fund called Malachite Capital Management in 2020, the bank established a Counterparty Oversight Committee (CPOC) specifically to address counterparty risk. A committee was created. Ownership was assigned. The governance response was, by any reasonable structural measure, appropriate.

CPOC identified issues and recommended solutions relating to Archegos as early as September 2020. It drew up actions for the Credit Risk Management unit to monitor for changes in the risk profile of Archegos. However, it failed to establish deadlines or assign ownership for those actions. As a result, substantive measures were not taken, and Archegos was not discussed at committee level again until March 2021 by which point the losses were already materialising.

The committee existed. The issues were identified. The actions were recommended. What was absent was the thing the slide could not show: whether anyone actually owned the outcome of those recommendations, in a way that was specific, time-bound, and consequential.

Three Ways Ownership Disappears

The gap between documented accountability and functional accountability does not appear all at once. It opens gradually, through mechanisms that are individually defensible and collectively corrosive.

The first is the authority gap.

Ownership without authority is not ownership - it is exposure. And yet most accountability frameworks assign responsibility for outcomes to people who do not control the full set of levers that produce them.

At Credit Suisse, this gap was explicit and documented. The bank's risk managers intended to ask Archegos for additional margin to reflect increased credit risk, but were prevented from doing so - this was deemed not to be in the interests of the bank by those responsible for managing the client relationship. The people accountable for risk management did not have the authority to exercise it when it conflicted with revenue priorities. The framework assigned them ownership of the outcome. The structure ensured they could not determine it.

This is not a Credit Suisse anomaly. It is the authority gap in its most transparent form and in most organisations, it operates with considerably less visibility.

The second is the diffusion problem.

Accountability, once shared, becomes negotiable. When ownership is distributed across a committee, the threshold for any individual to act unilaterally is permanently set just above what the situation appears to require.

Credit Suisse's CEO at the time stated he was not even aware of the existence of Archegos - despite the bank's exposure to the fund being four times that of its next largest client. This was not presented as negligence. It was presented as a structural reality - the information had not travelled up through the appropriate channels in the appropriate form. The governance model had distributed ownership so effectively that the person nominally most accountable had no functional connection to the risk being taken in the organisation's name.

Organisations have developed sophisticated language for this condition. They call it collective ownership, shared accountability, and matrix responsibility. These phrases describe real and sometimes necessary arrangements. They also describe the most reliable mechanism available for ensuring that, in practice, nobody owns anything.

The third is the escalation substitution.

In a well-functioning accountability structure, problems are resolved by the person closest to them, with escalation reserved for genuinely novel issues. In practice, escalation has become the primary mechanism for managing accountability rather than exercising it.

Credit Suisse's internal review found that concerns had been raised internally regarding the effectiveness of the Prime Services Risk Team - specifically a lack of staffing, failure to replace senior staff, and a lack of risk management experience in its leadership. These concerns travelled through the appropriate channels. They were noted, reviewed, and documented. The result was a 40% decrease in managing directors since 2019, the majority of whom left involuntarily as a result of cost cuts - leaving resources at less senior levels responsible for managing the risks their more experienced predecessors had previously owned.

The escalation path functioned. The resolution did not. The process demonstrated a considerably stronger capability for identifying problems than for fixing them, which is a distinction most governance frameworks are not designed to measure.

What the Slides Cannot Show

Marlene Amstad, chair of FINMA, noted after the collapse that Credit Suisse had "a cultural problem that translated into a lack of accountability." This is accurate but incomplete. Culture is the visible symptom. The structural design is the underlying condition.

The accountability framework, however well constructed, cannot capture the thing that actually determines whether ownership is real. It cannot show whether the named owner has the authority to make the decisions that matter. It cannot show whether escalation leads to resolution or recirculation. It cannot show whether the metrics being reported are the ones that reveal whether the process is working, or the ones that make it look as though it is.

The internal review into Archegos concluded that the design of Credit Suisse's risk management framework was effective. The framework was sound. The slide was correct. What failed was the translation of documented accountability into functional authority and that failure was not visible in any governance review until the losses made it impossible to ignore.

This is the core of the problem. Organisations have become skilled at designing accountability and considerably less skilled at testing whether it exists. The governance audit confirms that the right people are named and the right processes are in place. It does not typically ask whether those people have ever exercised the authority the model assumes they have.

The Comfort of the Slide

It is worth being honest about why this persists.

Performative accountability is not maintained through negligence. It is maintained because it is genuinely more comfortable than the alternative - for the organisation, for the people within it, and for the governance structures designed to oversee it.

Credit Suisse did not fail to reform. It reformed repeatedly. António Horta-Osório was brought in as chairman specifically to address governance and culture, with a reputation for successful institutional turnarounds. He described Credit Suisse's problems as worse than anything he had experienced in a three-and-a-half decade career. New frameworks were produced. New committees were established. New ownership structures were documented and signed off.

He resigned after nine months, having twice breached Covid restrictions. The underlying structural problems - the authority gaps, the diffusion of ownership, the substitution of escalation for resolution - remained intact beneath successive layers of governance reform.

Real accountability is disruptive. It requires someone to make decisions that inconvenience others. It requires metrics to be challenged and sometimes retired. It requires cross-functional dependencies to be renegotiated and institutional assumptions to be questioned.

Performative accountability requires none of these things. It requires attendance, documentation, and the maintenance of a governance structure that looks, from any reasonable external vantage point, exactly like the real thing. It is less disruptive, easier to sustain, and - crucially - it distributes the discomfort of accountability so effectively that no individual is ever required to bear enough of it to be motivated to change anything.

This is not a moral failure. It is a structural one. And structural failures require structural responses.

What Functional Accountability Actually Requires

The organisations that make accountability work do not have better frameworks. They have made different decisions about what ownership means and what it requires.

The first decision is about authority. Accountability must be accompanied by meaningful control over the outcomes it covers. The boundaries of authority must be explicit - the decisions an owner can make without negotiation clearly defined, and the cases requiring escalation the exception rather than the operating model. Where a genuine authority gap exists, that gap should be named and resolved, not documented and managed.

At Credit Suisse, risk managers were prevented from exercising the authority the framework assumed they had, because that authority conflicted with revenue priorities that sat outside their control. The fix was not a better risk framework. It was a decision about which took precedence - and who owned the consequences of that decision. That decision was never made explicitly, which meant it was made implicitly, repeatedly, in favour of revenue, until the accumulated consequences made the choice visible.

The second decision is about metrics. Accountability is only as meaningful as the outcomes it is measured against. FINMA's post-crisis report revealed that Credit Suisse met regulatory capital and liquidity requirements throughout the period of its decline. The metrics were clean. The slides were accurate. The institution was failing. The measures being reported were not the ones that would have revealed whether the accountability structures were functional - they were the ones that confirmed the framework was in place.

The third decision is about consequences. Accountability without consequence is a label. The Swiss government's attempt to claw back bonuses from Credit Suisse executives in 2023 was ruled unlawful in 2025, citing Swiss constitutional protections. The legal mechanisms for making the consequence of ownership real were not available after the fact. They needed to be designed in before the failure - which is precisely when designing them is most uncomfortable and most necessary.

The fourth decision is about honesty. The most important governance question most organisations are not asking is simple: does the person named as accountable actually own this - in the sense that they are making decisions, bearing consequences, and changing things that need to change? If the answer is no, the framework does not need refining. It needs replacing.

The Definitive Test

Ownership is not what appears on the slide, but what remains when the slide is closed.

It is the decision made when consensus was absent. The metric challenged when reporting it would have been easier. The cross-functional conversation initiated before it became unavoidable. The problem resolved rather than escalated.

Credit Suisse had governance infrastructure that satisfied every reasonable external test of accountability. It had committees, frameworks, ownership structures, and reform programmes. What it did not have - at the level that mattered - was a consistent answer to a deceptively simple question: who would still be standing next to this if it failed, and do they have enough authority to prevent it from doing so?

The forced merger with UBS raised important questions about accountability, market discipline, and the long-term risks taken within the banking sector. The disappearance of a bank of such scale and history underscores the reality that reputation, once damaged beyond repair, can outweigh even substantial capital buffers.

The lesson is not that Credit Suisse was uniquely poorly governed. The lesson is that governance which exists primarily as documentation - however thorough, however well-intentioned - is not a substitute for the thing it is designed to represent.

Most organisations will not face consequences at the scale Credit Suisse did. But most organisations are running the same structural risk, at a smaller magnitude, every time they fill in the ownership box and move on.

The process doesn't need another governance review. It needs someone who actually owns it - with the authority, the metrics, and the consequences to prove it.

What This Means in Practice

If ownership exists only on slides, the problem is not awareness, but design. And design can be changed - not through another governance review, but through a different set of decisions. Ones that are harder to approve and harder to ignore.

The Credit Suisse story is useful precisely because the failure was structural, not exceptional. Which means the vulnerabilities it exposed are worth testing for before the losses make them visible.

1. Test ownership against authority, not documentation

Does your named owner have uncontested authority over the outcomes they're accountable for? Take one "owned" process and ask: can this person make the decisions required to change the outcome without negotiation? If their authority can be overridden by revenue, relationship, or seniority without a formal resolution process, you have an authority gap. Name it or close it.

2. Replace shared accountability with explicit decision rights

When ownership is shared, who moves first? Committees can inform. They cannot own. Where multiple functions are involved, define who decides. If escalation is the default path, ownership has already dissolved. Identify your most consequential shared-ownership arrangements and ask: if this goes wrong at 2am, who calls whom, and does that person have the authority to act?

3. Introduce time-bound ownership as a minimum standard

Can you trace a recommendation to a resolution? No action without a deadline. No deadline without a named individual. No individual without consequence. If a recommendation can exist without all three, it is not an action. It is a note -and the Credit Suisse record is full of them.

4. Audit outcomes, not frameworks

Are your metrics revealing or reassuring? Capital adequacy and liquidity looked fine at Credit Suisse until they didn't. Stop asking: is the governance in place? Start asking: did this process produce the intended outcome under pressure? If the answer is unclear, the metrics are wrong - not the reporting cadence.

5. Make consequences real before they are needed

What happens to the person who owns a failure - not in policy, but in practice? If the honest answer is "they escalate, document, and survive," accountability is optional. Accountability that only activates after failure is not accountability. It is post-mortem analysis. Define upfront what happens when ownership fails. If that conversation feels uncomfortable, it is probably overdue.

The only question that matters

Most organisations do not lack accountability frameworks. They lack the willingness to make accountability operationally inconvenient - because real ownership forces trade-offs. Between risk and revenue. Between speed and control. Between alignment and decision. Those trade-offs cannot be resolved on a slide.

So the question is not "Who owns this?", but "Who can change this - and what happens if they don't?".

Until that question has a clear, enforceable answer, the ownership box is just formatting.

Ready? Let's talk.