Credit Suisse — risk, culture and the 2023 emergency merger

1. Episode summary

Credit Suisse Group AG, a 167-year-old Swiss global systemically important bank, entered the 2019–2023 window carrying a thinning capital buffer, a chronically loss-making investment bank and an accumulating record of compliance and conduct events. A 2019 corporate-surveillance affair — the bank's retention of private investigators to follow a departing senior executive — cost its then-CEO his position in 2020 and drew a formal reprimand from FINMA. In March 2021 two counterparty events landed inside three weeks: the collapse of the USD 10 billion group of supply-chain-finance funds that Credit Suisse ran with Greensill Capital, and the default of the Archegos family office, which left the prime-services business with roughly USD 5.5 billion in losses. An independent external investigation commissioned by the board (published July 2021) and later enforcement rulings by the Swiss Financial Market Supervisory Authority on Greensill (February 2023) and Archegos (July 2023) concluded that risks were visible in internal reporting well before each event and that the bank's first- and second-line controls failed to act. A second chairman departed in early 2022; a 2022 strategic restructuring and capital raise did not restore confidence; client outflows accelerated in Q4 2022. In March 2023, after a further confidence shock, the Swiss Confederation, the Swiss National Bank and FINMA brokered an emergency takeover by UBS announced on 19 March 2023. The strategic question the episode turned on was whether a G-SIB that repeatedly identified its own risk limits being breached could install the governance and execution to act on those signals before market confidence dissolved.

2. Sources

Primary:

1. Paul, Weiss, Rifkind, Wharton & Garrison LLP, "Report of the Special Committee of the Board of Directors of Credit Suisse Group AG – Archegos Capital Management," filed via Credit Suisse Form 6-K, 29 July 2021. (Full text of independent external investigation; 80+ interviews, 10+ million documents.)
2. Swiss Financial Market Supervisory Authority (FINMA), "FINMA concludes 'Greensill' proceedings against Credit Suisse," press release and enforcement ruling summary, 28 February 2023.
3. Swiss Financial Market Supervisory Authority (FINMA), "FINMA closes 'Archegos' proceedings against Credit Suisse," press release, 24 July 2023.
4. Swiss Financial Market Supervisory Authority (FINMA), "Lessons Learned from the CS Crisis" report, Bern, 19 December 2023 (covers 2008–2023 supervisory history).
5. Swiss Parliament, Parliamentary Investigation Commission (PInC/PUK), "Federal Authorities' Management of the Credit Suisse Crisis" report, 20 December 2024.
6. FINMA and UBS Group AG joint and separate press releases, "FINMA approves merger of UBS and Credit Suisse" and "UBS to acquire Credit Suisse," 19 March 2023.

Secondary (with justification):

1. Reuters, "Spies, lies and a chairman's exit: Credit Suisse's scandals," 17 January 2022. (Investigative wire synthesis of the 2019 surveillance affair and subsequent governance turnover.)
2. CNBC, "From spying to Swiss bailout: How years of turbulence at Credit Suisse came to a head," 17 March 2023. (Retrospective timeline of the 2019–2023 arc drawing on contemporaneous reporting.)
3. Morningstar, "Credit Suisse's Demise: A Timeline of Scandal and Failures," 2023. (Aggregated chronology compiled from filings and wire reports.)
4. M. Böni, "The UBS–Credit Suisse Merger: Helvetia's Gift," Journal of Applied Corporate Finance, 2025. (Peer-reviewed academic analysis of the rescue mechanics and capital position in the run-up.)
5. S&P Global Market Intelligence, "No strategy shift expected as Gottstein takes the reins at Credit Suisse," 6 February 2020. (Contemporaneous market analysis characterising Gottstein's appointment as continuity rather than strategic reorientation; used for Direction thin-evidence finding.)
6. CNBC, "Credit Suisse shares plunge 18% as bank announces huge third-quarter loss and strategic overhaul," 27 October 2022. (Wire record of the October 2022 CHF 4bn capital raise and CS First Boston restructuring announcement under Koerner; used for Direction and Act evidence.)
7. CNBC, "Credit Suisse posts massive annual loss as radical restructure gets underway," 9 February 2023. (Confirms 2022 full-year loss context and Q4 2022 client outflow acceleration; used for Act and Direction evidence.)
8. S&P Global Market Intelligence, "Credit Suisse CEO says he was not aware of Archegos issue before it made news," April 2021. (Gottstein's public statement that the executive board was not informed of the Archegos position prior to the default; used for Structure evidence.)

Tertiary (flagged):

1. BDO UK, "Credit Suisse and the Archegos collapse – lessons in Risk Management and Governance for all," practitioner commentary, 2021. (Flagged tertiary; used for frame only.)

3. OTA narrative

Observe. The observation apparatus was broadly adequate and the observation activity largely took place. The Paul Weiss report records that by mid-July 2020 the Archegos position was already in breach of the investment bank's scenario-exposure limits, and that the position remained in breach virtually every week from that point until the March 2021 default — limit reports, counterparty-oversight-committee papers, and risk-management escalations identified the concentration, under-margining and volatility of the book. The Greensill supply-chain-finance funds carried comparable visibility: FINMA's 2023 ruling states that concerns about the changing risk character of the funds, and about the concentration of obligors, reached the bank through internal channels and external warnings before the funds were gated. The FINMA 2023 lessons-learned report and the 2024 Parliamentary Investigation Commission report both conclude that the supervisor's own surveillance had identified the bank's destabilisation risk well before March 2023. Observation was routine for the G-SIB peer group: these were the standard reports a bank of this archetype produces and reads. Observe was not a root cause; it was the transmission step that carried correct signals into a decision layer that did not act on them.

Think. The reasoning step is where the case's weight sits. The independent investigation found that the risks visible in Observe were not translated into timely, binding decisions: risk-management committees deferred actions, deadlines lapsed without re-escalation, and a business orientation toward short-term revenue was permitted to override repeated limit breaches. FINMA's Greensill ruling is explicit that the bank relied on the counterparty itself to answer the critical questions its own reasoning should have resolved. The PInC report places primary responsibility on the board and executive board, noting they had "defied numerous interventions by FINMA in the preceding years" and had not translated documented concerns into corrective governance action. The interpretive frameworks required — counterparty-concentration analysis, scenario escalation, committee ownership of remediation — existed and were in standard use across the G-SIB peer group; the failure was that the reasoning from observation to decision was not performed with the rigour the evidence demanded. Think is a root-cause phase in this episode, classified as Easy-Wrong: the correct frameworks existed and were accessible to a reasonably-resourced G-SIB; they were not applied. The failure repeats across the surveillance affair, Greensill, Archegos and the 2022 restructuring — a pattern of recurrence the FINMA and PInC reports both characterise as an organisation that could not convert its own risk reads into decisions.

Act. Execution was mixed and, for part of the episode, a root-cause contributor. On the specific Archegos unwind of March 2021 the Paul Weiss report describes operational failures — the SEC parallel record and FINMA ruling likewise identify margining and unwind-sequencing shortcomings that widened the loss. Across the broader arc, the 2021 chairman transition, the 2022 strategic plan and capital raise, and the October 2022 restructuring were all actions taken by the bank; they did not stabilise confidence, and by Q4 2022 client outflows had accelerated to a level FINMA's 2023 report identifies as materially impairing the bank's recoverability. By March 2023 execution was operating under external constraints the reasoning failures of prior years had already set: the emergency takeover was executed competently by the authorities within a compressed weekend. Act is a root-cause phase for the specific Archegos unwind and for the multi-year failure to execute announced strategic restructurings; it is classified at the easy end of the task-difficulty axis — the routine risk-unwind and the routine follow-through on announced strategy were the standard moves for a G-SIB and were not performed to standard. Where Act was not a root cause — the March 2023 rescue mechanics — it was execution under external constraints the earlier Think failures had already produced.

4. Modality evidence

Direction. Direction evidence in this case is thin by the methodology's admissibility standard: the episode contains repeated strategic announcements but no single, dated, attributable choice that set a viable trajectory and was then pursued. Gottstein's appointment in February 2020 — following Tidjane Thiam's forced departure — was explicitly positioned as continuity rather than strategic reorientation; S&P Global Market Intelligence contemporaneous reporting described no strategy shift expected (Reuters/S&P Global synthesis, 2022). The October 2022 announcement under Ulrich Koerner of a "radical restructuring" — splitting the investment bank into a separate CS First Boston entity, raising CHF 4 billion in capital, and cutting risk-weighted assets by 40% — was the most concrete strategic direction signal of the episode (CNBC, "Credit Suisse shares plunge 18%," 27 October 2022). It met the Direction Evidence Rule's specificity and attribution criteria and was sufficiently dated. However, it did not survive long enough to constitute the trajectory of the episode: by Q4 2022, client outflows had accelerated, and by March 2023 the emergency UBS merger ended the bank's independent existence (FINMA, "Lessons Learned from the CS Crisis," December 2023; PInC report, 20 December 2024). The pattern across the four years is one of directional drift — no strategic pivot that committed the organisation to a specific future and held — rather than a single wrong direction deliberately chosen and executed. The Paul Weiss report and the PInC report both frame the failure as an organisation that could not translate its own risk reads into decisions, which is a Think and Process failure rather than primarily a Direction failure. Direction is present as context — the continued ambition to operate a full-service global investment bank on a shrinking capital base — but the admissibility bar for Direction as a primary modality is not met: no qualifying choice produced the outcome.

Scoring note (zero-modality rationale): the directional layer described in this subsection does not meet the methodology §5 Direction Evidence Rule three-prong admissibility test (specificity / timing / attribution) — the §4 evidence characterises the directional posture as direction evidence in this case is thin, not as a discrete, datable, attributable strategic choice. Direction is therefore inadmissible as a weight-carrying modality and is recorded at zero per cent; residual weight is redistributed across the other evidenced modalities (Structure, Processes, Culture) per methodology §3 redistribution formula. Categorisation under METHODOLOGY-ota-scoring-v4.md §5: modality acknowledged in narrative but not load-bearing — Direction Evidence Rule grounding.

Structure. The structural evidence is among the clearest in the case and is directly supported by primary sources. The Paul Weiss report identifies a structural gap in the governance of risky counterparty relationships: despite the existence of an executive board, there was no requirement that responsible executive board members address significant and risky business relationships on their own initiative as standard, and CEO Gottstein stated publicly he was not informed of the Archegos position before it made news (Paul Weiss report, July 2021; S&P Global Market Intelligence, April 2021). The Counterparty Policy and Oversight Committee (CPOC) — Credit Suisse's second-line governance mechanism specifically created in 2020 after the Malachite Capital Management losses — illustrates the structural failure precisely: CPOC identified Archegos exposure and recommended remediation as early as September 2020, but failed to establish deadlines and assign ownership for the actions it recommended, with the result that the committee did not discuss Archegos again at committee level until March 2021, the week the default occurred (Paul Weiss report, cited in BDO UK practitioner commentary, 2021). Reporting lines for the prime-services business did not surface the concentration to the executive board in a form that triggered mandatory action. FINMA's Greensill ruling (February 2023) made parallel structural findings: the bank "relied on the counterparty itself" to answer questions its own governance should have resolved, indicating a structural absence of independent second-line challenge authority in the asset-management business. The FINMA "Lessons Learned" report (December 2023) identified that between 2018 and 2022 alone FINMA carried out 108 on-site inspections and identified 382 points requiring action — a supervisory intensity that signals persistent structural remediation failure rather than occasional process lapses. The PInC report (December 2024) placed primary responsibility on the board and executive board, noting they had "defied numerous interventions by FINMA in the preceding years." Governance structure — specifically who held authority to force remediation of identified risk, and whether escalation paths from first and second lines reached the right decision-makers with binding consequence — is the primary structural failure mechanism.

Processes. Processes is the modality where the failure mechanism is most granular and most documented across primary sources. Three operational process failures are distinguished from structural failures using the methodology's Galbraith test: authority was nominally in place (the CPOC existed; limit reports were produced; the executive board had governance responsibility); what was missing was the machinery for exercising that authority with accountability and consequence. First, the limit-breach monitoring process: the Paul Weiss report records that the Archegos position was in breach of the investment bank's scenario-exposure limits by mid-July 2020 and remained in breach virtually every week until the March 2021 default — but limit reports did not trigger a mandatory escalation-and-remediation sequence with set deadlines and named owners. The CPOC's September 2020 identification of the problem and subsequent failure to re-table it until the default week is a process failure: a committee cycle with no closing mechanism allowed an acknowledged risk to sit unaddressed for six months. Second, the Archegos unwind process itself: the Paul Weiss report and the FINMA Archegos ruling (July 2023) both identify margining and unwind-sequencing shortcomings in March 2021 that widened the loss relative to what a correctly-sequenced unwind would have produced — an operational execution failure distinct from the earlier limit-monitoring failure. Third, the Greensill process failure documented in FINMA's February 2023 ruling: the bank's asset-management process for monitoring the changing risk character of the supply-chain-finance funds — obligor concentration, counterparty dependency, fund structure — did not produce a binding decision to gate or wind down the funds before the problem became external. The pattern across all three is the same: process machinery that produced monitoring output but lacked the closing-action architecture to convert that output into remediation.

Capability. The capability picture in this case is mixed and requires care to distinguish from the Process failure. The Paul Weiss report makes clear that the diagnostic capability to identify the Archegos risk was present: limit reports, counterparty-oversight-committee papers, and scenario-exposure calculations all produced accurate signals about the position's concentration, under-margining, and volatility (Paul Weiss report, July 2021). The capability gap was not in risk identification — the bank could see the exposure — but in a more specific institutional competence: translating repeated identified-limit breaches, in a prime-services context, into time-bounded management escalation with ownership and consequence. This is a narrower capability gap than the absence of risk-management competence altogether; it is closer to what the methodology distinguishes as institutional knowledge that did not survive the leadership churn of the period. Credit Suisse cycled through three chairmen (Rohner until April 2021, António Horta-Osório until January 2022, Axel Lehmann from January 2022), two CEOs (Thiam until February 2020, Gottstein until July 2022, Koerner from July 2022), and significant executive board turnover (Reuters synthesis, January 2022; CNBC retrospective, March 2023). The FINMA "Lessons Learned" report explicitly identifies that deficiencies in risk management identified by FINMA were "never sustainably remedied" — the word "sustainably" is significant, because it implies the capacity existed episodically but was not institutionalised. A reasonably-resourced G-SIB peer — Deutsche Bank's contemporaneous remediation programme, or HSBC's consent-order execution discipline — had established processes for tracking and closing regulator-identified remediation items in a documented, auditable sequence; the capability gap at Credit Suisse was the absence of that institutionalised discipline. Capability is a secondary contributor but is subordinate to Processes as the primary mechanism, because the evidence indicates the operational routines — not individual competences — were the missing element.

Scoring note (zero-modality rationale): the capability described in this subsection is recorded at zero per cent in the modality weights on the rationale of insufficient causal weight — the §4 evidence establishes that Credit Suisse possessed the technical and operational capability the situation required; the failure mechanism was located in Structure, Processes, Culture rather than in a capability gap. The capability is acknowledged as present in the narrative but does not carry standalone weight in the failure attribution. Categorisation under METHODOLOGY-ota-scoring-v4.md §5 "Zero-modality rationale rule": insufficient causal weight.

Culture. The cultural evidence in this case is extensive across primary and secondary sources and characterises a normative environment in which revenue-and-reputation priorities suppressed risk escalation and regulator responsiveness over the full episode length. The 2019 surveillance affair is the earliest and most vivid cultural signal: COO Pierre-Olivier Bouée ordered private investigators to follow departing head of wealth management Iqbal Khan through Zurich without informing CEO Thiam or the board, in order — per the investigation — "to protect the interests of the bank" (Reuters, 17 January 2022; CNBC retrospective, 17 March 2023). The affair's cultural significance lies not in the specific act but in what it reveals about the normative environment: a C-suite executive committing reputational resources to monitoring a competitor recruit rather than to the risk governance failures that would materialise two years later. Thiam's departure in February 2020 was itself forced by board pressure following the surveillance affair, not by strategic underperformance, which indicates a board culture that could remove a CEO over a visibility-and-optics event but did not remove leadership over accumulating, documented risk-limit breaches. FINMA Chair Marlene Amstad characterised the failure explicitly: Credit Suisse had a "cultural problem that translated into a lack of accountability" (Risk Academy summary, citing Amstad). The PInC report (December 2024) is the most direct primary-source characterisation of the cultural dynamic: the board and executive board had "defied numerous interventions by FINMA in the preceding years," and FINMA found Credit Suisse's managers had been "reticent" — in FINMA's own words, the polite description — when the regulator intervened on remuneration and governance matters. The FINMA "Lessons Learned" report (December 2023) identifies that despite 382 FINMA-identified action points across 108 inspections, the deficiencies "were never sustainably remedied," consistent with an organisation whose cultural defaults prioritised the appearance of compliance over the institutional commitment to execute remediation. The normalisation of this posture across surveillance affair, Greensill, Archegos, and the Q4 2022 confidence collapse — four qualitatively distinct failure events, each preceded by documented internal signals, each failing to produce binding pre-emptive corrective action — is the hallmark of an organisation whose cultural norms prevented the conversion of observed risk into action. This is the Culture evidence pattern: the formal machinery (limit reports, CPOC, FINMA proceedings) existed, but the normative environment prevented people from engaging with it honestly and with consequence. The methodology's Processes/Culture test — "Is the problem in the written machinery, or in whether people engage with it honestly?" — resolves here to Culture as the upstream factor: the CPOC process existed; people on it declined to assign ownership and deadlines.

---