DOI: 10.1111/jacf.12674 O R I G I N A L A RT I C L E The UBS-Credit Suisse Merger: Helvetia’s Gift Pascal Böni1 Tim A. Kroencke2 Florin P. Vasvari3 1Tilburg University, Tilburg, The Netherlands 2FHNW School of Business, Basel, Switzerland 3London Business School, London, UK Correspondence Pascal Böni, Tilburg University, The Netherlands. Email: pascal.boeni@tilburguniversity.edu INTRODUCTION Pietro Veronesi and Luigi Zingales provide an account of the staggering costs of extensive government intervention in the US financial sector during the 2008 global financial crisis.1 To reduce such costs in the future, extensive regulation has been introduced to make banks more resilient, and to protect taxpayers and private investors from bearing bailout costs.2 But a key question remains: Is the post-2008 regulatory framework effective? In this paper, we analyze the UBS-Credit Suisse merger to shed light on this question. On the evening of Sunday, March 19, 2023, the Swiss Fed- eral Council, the Swiss National Bank, and the Swiss Financial Market Supervisory Authority (Finma) jointly announced the orchestrated bailout-merger of Credit Suisse (CS) by its domes- tic banking rival UBS Group AG (UBS), marking the end of 167 years of proud Swiss banking history.3 The demise of CS shook faith in a stable Swiss Confederation, often affectionately called “Helvetia”.4 The bailout-merger, which aimed to restore confidence in the Swiss financial system, deviated significantly from standard bank resolution procedures. It lacked competitive bidding and circum- vented a typical bank resolution or purchase and assumption (P&A) transaction, where the acquiring bank purchases the failed 1 Veronesi, P., and L. Zingales. 2010. “Paulson’s Gift.” Journal of Financial Economics 97: 339- 368. 2 Thakor, A.V. 2015. “The Financial Crisis of 2007–2009: Why Did It Happen and What Did We Learn?” Review of Corporate Finance Studies 4: 155–205. 3 We prefer to use the term “bailout”, rather than “rescue” or “lifeboat operation” to emphasize the use of emergency powers by the government (the Federal Council) and the significant guarantees provided by taxpayers to UBS, which posed a substantial risk to public funds. Therefore, we refer to the transaction as a “bailout-merger” as it involved a state-orchestrated merger with the primary goal of bailing out CS. 4 Helvetia is the female national personification of Switzerland (Confederatio Helvetica), often pictured in a flowing gown, with a spear and a shield with the Swiss flag. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2025 The Author(s). Journal of Applied Corporate Finance published by Wiley Periodicals LLC on behalf of Cantillon & Mann. bank’s assets and assumes its deposits. Instead, the Swiss gov- ernment forced the implementation of a government emergency rescue deal, which consisted of a complete emergency merger share-deal between UBS and CS. This emergency rescue deal also included massive state liquidity guarantees in the amount of 214 billion (bn) US dollars (USD) and, additionally, a sub- stantial loss guarantee totaling 9.63 bn USD to cover potential losses incurred on the realization of certain CS assets. We argue that the exclusion of competitive bidding, imposed by the gov- ernment, and the relatively late intervention of the regulator have led to an unexpectedly favorable deal for the acquirer, UBS. We show that significant wealth transfers to specific asset owners have taken place due to the merger. While some of these wealth trans- fers were offset by redistributions from CS shareholders and AT1 bondholders, the ones who are supposed to bear the burden of bankruptcy, the overall wealth effect cannot be solely explained by the participating firms’ abnormal returns on securities. We provide insights into the merger-induced value creation and destruction and the redistribution of wealth amongst stakeholders and tax- payers. More specifically, we show that Switzerland’s cost of debt increased substantially as a consequence of the state-orchestrated merger between UBS and CS. We conclude that an economically meaningful part of the costs is borne exogenously, that is, primar- ily by the taxpayer. This is what we call the “Helvetia’s gift”, which suggests that the current regulatory framework does not actually protect the public from bad behavior by financial actors as much as one might hope. To reach this conclusion, we undertake three steps. First, we quantify the wealth effects for the stockholders and bondholders of UBS (acquiror) and CS (target). Second, we compare our empir- ical findings with insights from extant academic literature on competitive bank mergers. Third, we assess the anticipated refi- nancing cost of the massive liquidity and loss guarantees granted by the Swiss government. 104 wileyonlinelibrary.com/journal/jacf J. Appl. Corp. Finance. 2025;37:104–121. mailto:pascal.boeni@tilburguniversity.edu http://creativecommons.org/licenses/by/4.0/ https://wileyonlinelibrary.com/journal/jacf http://crossmark.crossref.org/dialog/?doi=10.1111%2Fjacf.12674&domain=pdf&date_stamp=2025-07-06 105 We estimate stockholder wealth effects using high-frequency intraday stock data over the period from Friday, March 17 (5:30 p.m.) to Tuesday, March 21 (5:30 p.m.). The bailout-merger resulted in a 2-day cumulative abnormal return (CAR) of 7.95% for UBS shareholders and a −55% CAR for CS shareholders, while other European banks show no significant abnormal stock returns.5 In absolute values, relative to their market capitalizations as of March 17, 2023, these abnormal returns translate to a wealth increase of 5.14 bn USD for UBS stockholders, and a wealth decrease of 4.35 bn USD for CS stockholders,6 with a dispro- portionate negative impact on small equity retail investors in CS, as opposed to large institutional investors.7 Therefore, we observe a positive combined stockholder wealth effect of approximately 0.79 bn USD. Next, we examine bondholder wealth effects resulting from the merger-bailout. The deal involved the write-down of AT1 bonds with a nominal value of 17 bn USD and an approximate market value of 3.9 bn USD.8 While the AT1-write-down and associated numbers have received extensive media coverage, less attention has been given to the impact of the merger on CS’s and UBS’s holders of straight bonds even though they had significantly higher value compared to AT1 bonds. Since bond markets are generally less liquid,9 we first analyze intraday high-frequency data from credit default swap (CDS) spreads. Price information derived from CDS spreads is based on informed price discovery by traders in a liquid market known for accurately trading credit risk.10 Our findings reveal economically substantial and statistically significant cumu- lative abnormal CDS spread changes (CAC) of −755 basis points (bp) for CS over the 2-day horizon. In contrast, UBS’s spread decreases by an insignificant 4 bp during the same period. The large abnormal CDS spread changes indicate that CS bondholders experienced significant abnormal returns since CDS spreads are sensitive to credit events and are closely related to yield spreads.11 To estimate bondholder wealth effects more accurately in USD, we utilize daily bond data for 57 CS bonds, which account for approximately 80% of CS’s long-term debt. The 2-day CAR for the (observable) value-weighted CS bond portfolio amounts to an impressive +34.74%. In absolute values, relative to the market value of the target’s bond portfolio as of March 17, 2023, these 5 During the same event-window, other European global systemically important banks (G- SIBs) show no significant CARs, indicating that these abnormal positive combined returns indeed reflect news specific to the bailout- merger rather than a global shift in investors’ perception of risk. 6 According to Bloomberg data, the market capitalization of UBS and CS on March 17, 2023, was 60.37 bn Swiss francs and 7.44 bn Swiss francs, respectively. To calculate the approximate wealth increase or decrease in USD, we use the exchange rate of 1.07. 7 According to Bloomberg data, institutional equity ownership in CS (as a percentage of free float) declined from 59.23% in mid-2022 to 39.64% just before the merger announcement. Concurrently, short interest positions increased from 11,310,009 shares shorted in July 2022 to 28,584,988 shares shorted in December 2022. These findings indicate that the bailout- merger has had a disproportionate negative impact on small equity retail investors in CS. 8 On the evening of March 17, 2023, the equally weighted portfolio of AT1 bonds, for which price information was available, traded at an approximate value of 22.9% of the face value. 9 See Edwards, A., L. Harris, and M. Piwowar. 2007. “Corporate bond market transparency and transaction costs.” Journal of Finance 50: 1421–1451; and Bessembinder, H., K.M. Kahle, W.F. Maxwell, and D. Xu. 2009. “Measuring Abnormal Bond Performance.” Review of Financial Studies 22: 4219–4258 10 See Veronesi and Zingales (2010). 11 Duffie, D. 1999. “Credit Swap Valuation.” Financial Analysts Journal 55: 73–87; Jorion, P., and G. Zhang. 2007. “Good and Bad Credit Contagion: Evidence From Credit Default Swaps.” Journal of Financial Economics 84: 860–883; and Blanco, R., S. Breannan, and I.W. Marsh. 2005. “An Empirical Analysis of the Dynamic Relation Between Investment-Grade Bonds and Credit Default Swaps.” Journal of Finance 60: 2255–228 abnormal returns correspond to a significant and economically important value-weighted bondholder wealth increase of 22.65 bn USD. At the same time, we find no wealth increase for UBS bond- holders. Accounting for the net AT1-bond wealth changes (−3.9 bn USD), these findings suggest a total wealth increase of 18.75 bn USD for CS’s bondholders. Therefore, considering the calculated total stockholder and bondholder wealth effects outlined above, the combined wealth increase amounts to 19.5 bn USD (0.79 bn USD net stockholder effects plus 18.75 bn USD net bondholder effects). This can be interpreted as the net market value created by the state orches- trated bailout-merger deal. The entire net wealth effect appears to be exogenous, not attributable to any wealth transfers from bondholders to stockholders within or across the merging banks. Could CS have been rescued at a lower cost? We posit that allowing for competitive bidding for CS’s equity would have likely resulted in a lower price for its rescue. While it is challenging to establish this quantitatively, we draw on comprehensive academic research on competitive merger bids to support this contention. First, prior literature finds negative stockholder abnormal returns for non-failed bank acquisitions12 and modest positive CARs for failed-bank acquisitions.13 These modest CARs are primarily attributed to bidder restrictions.14 In competitive bidding scenar- ios, the winning bidder often overpays, leading to more favorable terms of the target’s shareholders. Therefore, drawing on the win- ner’s curse hypothesis of Richard Roll15 and existing literature, we argue that bidder restrictions likely resulted in a wealth transfer from CS to UBS stockholders. Second, the literature suggests that the wealth transfer to CS bondholders may be attributed to a coinsurance effect.16 With the merger announcement, the market anticipated a substantial decrease in CS’s leverage and probability of default. It is evident that an unexpected decrease in firm leverage can lead to wealth 12 Bliss, R.T., and R. Rosen. 2001. “CEO Compensation and Bank Mergers.” Journal of Finan- cial Economics 61: 107–138; DeLong, G., and R. DeYoung. 2007. “Learning by Observing: Information Spillovers in the Execution and Valuation of Commercial Bank M&As.” Journal of Finance 62: 181–216; and Houston, J.F., C.M. James, and M.D. Ryngaert. 2001. “Where Do Merger Gains Come From? Bank Mergers From the Perspective of Insiders and Outsiders.” Journal of Financial Economics 60:, 285–331; and Sushka, M.E., and Y. Bendeck. 1988, “Bank Acquisition and Stockholders’ Wealth.” Journal of Banking & Finance 12: 551–562. 13 Baibirer, S.D., G.D. Jud, and F.W. Lindahl. 1992. “Regulation, Competition, and Abnor- mal Returns in the Market for Failed Thrifts.” Journal of Financial Economics 31: 107–131; Bertin, W.J, F. Ghazanfari, and K.M. Torabzadeh. 1989. “Failed Bank Acquisitions and Suc- cessful Bidders’ Returns.” Financial Management 18: 93–100; Cochran, B., L. Rose, and D.R. Fraser. 1995. “A Market Evaluation of FDIC Assisted Transactions.” Journal of Banking and Finance 19: 261–279; Cowan, A.R., and Salotti, V. 2015. “The Resolution of Failed Banks During the Crisis: Acquirer Performance and FDIC Guarantees, 2008 – 2013.” Journal of Banking & Finance 54: 222–238; James, C.M., and P. Wier. 1987. “Returns to Acquirers and Competition in the Acquisition Market: The Case of Banking.” Journal of Political Economy 95: 355–370; and Zhang (1997). 14 See Baibirer et al. (1992); Bertin, W.J, F. Ghazanfari, and K.M. Torabzadeh. 1989. “Failed Bank Acquisitions and Successful Bidders’ Returns.” Financial Management 18: 93–100; Cochran, B., L. Rose, and D.R. Fraser. 1995. “A Market Evaluation of FDIC Assisted Trans- actions.” Journal of Banking and Finance 19: 261–279; Cowan and Salotti (2015); Gilberto, M., and M. Varaiya. 1989. “The Winner’s Curse and Bidder Competition in Acquisitions: Evidence From Failed Bank Auctions.” Journal of Finance 44: 59–75; and James and Wier (1987). 15 Roll, R. 1986. “The Hubris Hypothesis of Corporate Takeovers.” Journal of Business 59: 197—216. 16 Billett, M.T., T.-H.D. King, and D.C. Mauer. 2004. “Bondholder Wealth Effects in Mergers and Acquisitions: New Evidence From the 1980s and 1990s.” Journal of Finance 59: 107– 135; Bodnaruk, A., and M. Rossi. 2016, “Dual Ownership, Returns, and Voting in Mergers.” Journal of Financial Economics 120: 58–80; and Chen, S-S., K.-Y. Ho, P.-H. Ho, and W.-Y. Nie. 2022. “CEO Overconfidence and Bondholder Wealth Effects: Evidence From Mergers and Acquisitions.” Journal of Corporate Finance 77: 1—27. 17456622, 2025, 2, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/jacf.12674 by Fachhochschule N ordw estschw eiz, W iley O nline L ibrary on [15/01/2026]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 106 JOURNAL OF APPLIED CORPORATE FINANCE transfers from stockholders to bondholders.17 This coinsurance effect is particularly pronounced when the target’s rating is lower than the acquirer’s or when the acquisition is expected to reduce the target’s risk.18 Both conditions were present in this merger, which supports the existence of large abnormal returns. How- ever, the significant wealth gain of almost 18.75 bn USD for CS bondholders, combined with no change in the value of UBS bonds, suggests that this mechanism alone cannot fully explain the observed effects. A third additional factor at play may be the “too-big-to-fail” channel whereby the new bank likely benefits from reinforced gains associated with its “too-big-to-fail” status.19 An important element of this takeover was the loss protection agreement signed by UBS with the Federal Department of Finance (FDF). This agreement covered a specific portfolio of Credit Suisse assets, which corresponded to approximately 3% of the combined assets of the merged bank. UBS could draw the guarantee for any real- ized losses exceeding CHF 5 bn from the federal government (up to a maximum of CHF 14 bn). Only losses realized could be covered by this guarantee. In support for this channel, we find that the government intervention resulted in a significant jump in Switzerland’s cost of debt, ultimately placing a burden on taxpay- ers. Consistent with the prior literature on the cost of government interventions the event caused a substantial increase in Switzer- land’s sovereign credit risk and, consequently, its expected cost of capital.20 Switzerland’s sovereign credit risk, as proxied by its CDS spread, more than doubled. The present value of the associ- ated expected increase in capital costs, amounts to approximately 5.8–7.2 bn USD. We thus conclude that the substantial combined net wealth increase of 19.5 bn USD, unexplained by abnormal security returns, ultimately falls on the shoulders of taxpayers. Both, the loss protection agreement mentioned above but also the observed jump in Switzerland’s cost of debt do support this interpretation. A poorly managed bank is kept afloat, and an incentive for large banks to take excessive risks and lower their efforts to manage risks is heightened. While these costs may be outweighed by ben- efits such reducing the likelihood of a financial panic, achieving these benefits at a lower cost should have been the primary goal. This could have been accomplished through the avoidance of bid- der restrictions and effective bank oversight that utilizes existing market signals in a timely manner to facilitate an orderly bank resolution. The subsequent sections of the paper proceed as follows. Sec- tion 2 provides a description of the events leading up to the UBS/CS bailout-merger, Section 3 outlines the data and event study methodology used, Section 4 presents the empirical results along with robustness tests, Section 5 discusses the findings, and Section 6 concludes the paper. 17 For example, Jorion and Zhang (2007) 18 Billet et al. (2004); Bodnaruk and Rossi (2016) 19 Penas, M.F., and H. Unal. 2004. “Gains in Bank Mergers: Evidence From the Bond Markets.” Journal of Financial Economics 74: 149–180. 20 Acharya, V., and T. Yorulmazer. 2007. “Cash-In-The-Market Pricing and Optimal Resolution of Bank Failures.” Review of Financial Studies 21: 2705—2742. PRELUDE TO THE UBS/CS—MERGER On March 19, 2023, the Swiss Financial Market Supervisory Authority (Finma) and the Swiss National Bank (SNB) jointly announced a state-orchestrated bailout-merger between UBS and CS. This bailout involved the wipeout of Additional Tier 1 (AT1) bonds with a nominal amount of 17 bn USD. The crisis stemmed from a loss of investor confidence in CS, one of Switzer- land’s largest banks, which had previously been known for its robust financial safeguards mandated by regulators. Several events preceded the bailout-merger: 1. The series of events began with a spying scandal in February 2020, leading to the abrupt departure of CS’s CEO, Tidjane Thiam. The bank had hired private detectives to spy on its former head of wealth management, Iqbal Khan, who had joined UBS. 2. In March 2021, CS faced pressure to close four funds connected to the collapse of Greensill Capital, a British financial firm specializing in short-term corporate loans in which around 10 bn USD had been invested. Finma claimed that CS had severely breached its supervisory obligations, resulting in four enforcement proceedings against former CS managers. 3. Also in March 2021, CS’s investment bank suffered a loss of 5.5 bn USD due to its involvement with Archegos Capital Management, a US hedge fund that went into default. The hedge fund held highly leveraged assets, primarily technology stocks, and CS had partially financed its activities.21 4. In October 2021, CS was fined 547 million USD by US and British authorities for its involvement in dealings related to 1.3 bn USD loans to state-owned companies in Mozam- bique. Allegedly, CS had paid 200 million USD in kickbacks to bankers and government officials in Mozambique.22 5. In January 2022, CS’s new chairman since May 2021, Anto- nio Horta-Osorio faced accusations of violating Switzerland’s Covid restrictions and subsequently resigned.23 He was hired from Lloyds Banking Group, where he was a CEO, to man- age CS’s turnaround and implement better risk management practices. 6. In February 2022, a global media investigation and data leak concerning more than 18,000 bank accounts revealed that CS had allegedly been involved in banking transac- tions with dubious legal entities and individuals, including heads of state, intelligence officials, drug lords, and sanc- tioned businessmen associated with serious crimes such as drug trafficking, corruption, and money laundering. The scandal again further tarnished the bank’s reputation.24 21 See the 165-page review by law firm Paul Weiss, Rifkind, Wharton, and Garrison for more details related to the Archegos default and CS’s involvement: https://www.credit-suisse.com/ about-us/en/reports-research/archegos-info-kit.html 22 See https://www.reuters.com/business/finance/spies-lies-regulators-round-credit-suisse- 2021-10-20/ for more information in respect to the Mozambique case. 23 See https://www.faz.net/aktuell/wirtschaft/unternehmen/credit-suisse-praesident-horta- osorio-geht-wegen-quarantaeneverstoss-17735433.html for more information on this. 24 See https://www.reuters.com/business/finance/swiss-prosecutors-launch-case-over-credit- suisse-dirty-money-data-leak-reports-2023-02-03/ for more information on the dirty money scandal. 17456622, 2025, 2, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/jacf.12674 by Fachhochschule N ordw estschw eiz, W iley O nline L ibrary on [15/01/2026]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense https://www.credit-suisse.com/about-us/en/reports-research/archegos-info-kit.html https://www.credit-suisse.com/about-us/en/reports-research/archegos-info-kit.html https://www.reuters.com/business/finance/spies-lies-regulators-round-credit-suisse-2021-10-20/ https://www.reuters.com/business/finance/spies-lies-regulators-round-credit-suisse-2021-10-20/ https://www.faz.net/aktuell/wirtschaft/unternehmen/credit-suisse-praesident-horta-osorio-geht-wegen-quarantaeneverstoss-17735433.html https://www.faz.net/aktuell/wirtschaft/unternehmen/credit-suisse-praesident-horta-osorio-geht-wegen-quarantaeneverstoss-17735433.html https://www.reuters.com/business/finance/swiss-prosecutors-launch-case-over-credit-suisse-dirty-money-data-leak-reports-2023-02-03/ https://www.reuters.com/business/finance/swiss-prosecutors-launch-case-over-credit-suisse-dirty-money-data-leak-reports-2023-02-03/ 107 7. In March 2022, a Bermuda judge ruled that CS Life Bermuda, CS’s local life insurance subsidiary, owed damages of 553 million USD to former Georgian prime minis- ter Bidzina Ivanishvili due to mismanagement. The fraud committed by a former CS banker, Patrice Lescaudron, who abused the trust of CS clients, also contributed to the rul- ing.25 Lescaudron was sentenced to five years in prison in 2018 and later committed suicide in 2020. 8. In June 2022 Switzerland’s Federal Criminal Court found CS and a former employee guilty of money laundering on behalf of a Bulgarian cocaine-trafficking ring. The illicit funds were allegedly laundered through CS accounts, resulting in a fine of 2 million Swiss francs (CHF).26 9. In October 2021, CS’s new Chairman, Axel Lehmann and CS’s new CEO, Ulrich Koerner, unveiled a plan to cut 9000 jobs and successfully raised 4.3 bn USD in fresh capital through a fully underwritten rights issue and a private cap- ital placement of 1.76 bn USD. The Saudi National Bank became CS’s largest shareholder as a result.27 The two also announced a plan to carve out the investment banking opera- tions and spin off the revived First Boston unit, the US-based investment bank it acquired in 1990.28 10. In the fourth quarter of 2022, CS suffered significant cus- tomer outflows of over 100 bn CHF (approximately 119 bn USD), leading to a record annual loss of 7.29 bn CHF, accounting for approximately 50% of CS’s net revenues!29 In response, longtime CS shareholder Harris Associates sold its entire stake in the bank.30 These events have likely damaged CS’s reputational capital, which matters in financial contracting.31 Not surprisingly, CS’s market capitalization lost an estimated 30 bn USD or 90% of its value over the period starting at the beginning of 2020 and end- ing with the bailout-merger. This decrease in value aligns with prior research which provides evidence that news about possible financial misconduct significantly and negatively impacts share prices.32 However, the events surrounding CS had not yet to reached their peak: 1. On March 8, 2023, CS delayed its annual report following a call from the US Securities and Exchange Commission (SEC), 25 See https://www.reuters.com/world/europe/former-georgian-pm-wins-bermuda-claim- against-credit-suisse-subsidiary-2022-03-29/ for more information on the Bidzina Ivanishvili scandal. 26 See https://www.reuters.com/business/finance/court-convicts-credit-suisse-money- laundering-case-2022-06-27/ for more information on the cocaine cash laundering case. 27 See https://www.bloomberg.com/professional/blog/credit-suisse-raises-4-3-bn-capital- after-wild-ride/ for details on the capital raise here. 28 See https://www.credit-suisse.com/media/assets/corporate/docs/about-us/media/media- release/2022/10/strategy-update-press-release-en.pdf for more details. 29 See https://www.credit-suisse.com/media/assets/corporate/docs/about-us/media/media- release/2023/02/q4-22-press-release-en.pdf for the 4th quarter 2022 results and CS 4Q22 and Full Year 2022 Results. 30 See Financial Times, March 5, 2023, “Former top Credit Suisse shareholder Harris Associates sells out of bank”. 31 Beatty, R.P, H. Bunsis, and J.R.M. Hand. 1998. “The Indirect Economic Penalties in SEC Investigations of Underwriters.” Journal of Financial Economics 50: 151-186; Fang, L.H. 2005. “Investment Bank Reputation and the Price and Quality of Underwriting Services.” Journal of Finance 60: 2729–2761; Atanasov, V., V. Ivanov, and K. Litvak. 2012, “Does Reputation Limit Opportunistic Behavior in the VC Industry? Evidence From Litigation Against VCs.” Journal of Finance 62: 2215–2246. 32 See Amiram, D., Z. Bozanic, J.D. Cox, Q. Dupont, J.M. Karpoff, and R. Sloan. 2018. “Financial Reporting Fraud and Other Forms of Misconduct: A Multidisciplinary Review of the Literature.” Review of Accounting Studies 23: 732–783 for a comprehensive review of this literature. which raised questions about revisions to cash flow statements from 2019 and 2020, as well as related controls. 2. On March 14, CS released its annual report admitting “mate- rial weaknesses” in its financial controls and announcing the elimination of board bonuses. 3. On March 15, 2023, the chairman of Saudi National Bank, responded with a blunt “absolutely not” when asked by a journalist if they would offer additional financial support to Credit Suisse if needed. This response added to the mounting concerns surrounding CS. 4. On March 16, in line with its legal mandate, the SNB provided a liquidity line of 50 bn Swiss francs. 5. On the weekend of March 18 and 19, the SNB, Finma and the Federal Council convened to address the crisis. They bro- kered a deal for UBS to acquire CS, aiming to stabilize Switzerland’s financial system. The Financial Times reported on March 18 that BlackRock was considering a rival bid for CS, but ultimately abandoned the idea as SNB and Finma favored a Swiss solution.33 6. On the evening of March 19, 2023, a press conference was held where the SNB, Finma and the Federal Council announced UBS’s agreement to acquire CS for 3.23 bn USD in stock, along with assuming up to 5.4 bn USD in losses. Additionally, AT1 bonds with a nominal amount of 17 bn USD were written off.34 To facilitate the bailout-merger, the Federal Council enacted emergency measures based on articles 184 and 185 of the Fed- eral Constitution. These measures included the creation of a legal framework allowing the national bank to provide additional liq- uidity assistance beyond standard emergency liquidity assistance. The Federal Council also provided a default guarantee to the SNB. The Finance Delegation, representing the federal government and driven by the Federal Council), granted a 9 bn Swiss franc guar- antee to cover potential losses arising from specific assets UBS acquired as part of the transaction.35 UBS was responsible for the first CHF 5 bn of any realized losses associated with winding down inherited Credit Suisse assets that were deemed non-core or 33 See FT, March 18, 2023: “BlackRock explored rival Credit Suisse takeover bid” 34 Together with common equity tier one (CET1) capital, Additional Tier 1 (AT1) bonds, also known as contingent convertibles or “CoCos,” act as a first line of defense to absorb losses. They serve to protect taxpayers, depositors, and senior debt investors from the financial consequences of a distress event. AT1 bonds are converted into equity or written off in specific scenarios, often related to the capital ratio of the issuing bank. They were introduced after the 2008 financial crisis, as regulators aimed to shift risk away from taxpayers and increase the capital reserves held by financial institutions to safeguard against future crises. If AT1 bonds are triggered in a distress scenario, bondholders may lose their entire investment, lose interest payments, or end up with equity holdings in a weakened bank. In the case of CS, the write-off of AT1 bonds was consistent with their prospectus, which explicitly stated that in the event of a “contingency event,” such as the Core Equity Tier One capital ratio falling below the required threshold or a “viability event,” “the full principal amount of each Note will be written down to zero,” and “the holders will be deemed to have irrevocably waived their rights to repayment of the aggregate principal amount of the Notes.” To remove any ambiguity regarding the status of Credit Suisse’s AT1 bonds, a legislative amendment was made on the day the deal was agreed upon: “At the time of the credit approval in accordance with Article 5, Finma may order the borrower and the financial group to write down additional Tier 1 capital” (Swiss Federal Council ordinance, March 19, 2023). The bonds contractually stipulate that they will be entirely written down if extraordinary government support is granted. The Finma argued that this condition was met and given that Credit Suisse “received extraordinary liquidity assistance loans secured by a federal default guarantee on 19 March.” However, various CS investors sue the Swiss regulator over this AT1 bond wipeout and claim that Finma failed to behave proportionately and in good faith. 35 See press release of the Federal Council of March 19, 2023: https://www.admin.ch/gov/en/ start/documentation/media-releases.msg-id-93793.html 17456622, 2025, 2, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/jacf.12674 by Fachhochschule N ordw estschw eiz, W iley O nline L ibrary on [15/01/2026]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense https://www.reuters.com/world/europe/former-georgian-pm-wins-bermuda-claim-against-credit-suisse-subsidiary-2022-03-29/ https://www.reuters.com/world/europe/former-georgian-pm-wins-bermuda-claim-against-credit-suisse-subsidiary-2022-03-29/ https://www.reuters.com/business/finance/court-convicts-credit-suisse-money-laundering-case-2022-06-27/ https://www.reuters.com/business/finance/court-convicts-credit-suisse-money-laundering-case-2022-06-27/ https://www.bloomberg.com/professional/blog/credit-suisse-raises-4-3-bn-capital-after-wild-ride/ https://www.bloomberg.com/professional/blog/credit-suisse-raises-4-3-bn-capital-after-wild-ride/ https://www.credit-suisse.com/media/assets/corporate/docs/about-us/media/media-release/2022/10/strategy-update-press-release-en.pdf https://www.credit-suisse.com/media/assets/corporate/docs/about-us/media/media-release/2022/10/strategy-update-press-release-en.pdf https://www.credit-suisse.com/media/assets/corporate/docs/about-us/media/media-release/2023/02/q4-22-press-release-en.pdf https://www.credit-suisse.com/media/assets/corporate/docs/about-us/media/media-release/2023/02/q4-22-press-release-en.pdf https://www.admin.ch/gov/en/start/documentation/media-releases.msg-id-93793.html https://www.admin.ch/gov/en/start/documentation/media-releases.msg-id-93793.html 108 JOURNAL OF APPLIED CORPORATE FINANCE incompatible with its risk profile. If losses exceed this amount, the federal government has committed to cover up to a maxi- mum of CHF 9 bn. This Swiss federal guarantee obliged UBS to manage the assets in such a way that losses are minimized (and realization proceeds are maximized) and the federal government received broad information and audit rights in order to verify this. Furthermore, CS and UBS received a total of 200 bn Swiss francs in additional liquidity assistance loans from the Swiss National Bank, comprising a 100 bn Swiss franc loan with privileged cred- itor status in bankruptcy and a loan of up to 100 bn Swiss francs backed by a federal default guarantee.36 As of the end of May 2023, Credit Suisse had repaid its outstanding liquidity amounts received in full to the Swiss National Bank. Almost a month later, on May 16, 2023, UBS disclosed poten- tial costs and benefits amounting to tens of bns of dollars from its takeover of CS, highlighting the significant stakes involved in completing the rescue of its struggling Swiss rival. UBS estimated a negative impact of $13 bn from fair value adjustments and $4 bn in potential litigation and regulatory costs resulting from outflows. Additionally, the switch in accounting standards brought the total hit to $28.3 bn. However, UBS expected to offset these costs with a write-down of $17.1 bn from Credit Suisse’s AT1 bonds as well as taking over CS for a fraction of its book value, resulting in a one-off gain of $34.8 bn from the acquisition. While the disclosure of the accounting gain was seen as less favorable than expected, it did offer UBS a cushion to absorb losses and costs associated with the merger and was likely to contribute to a boost in UBS’s future profits if the transaction pro- ceeded as planned. The numbers underscored CS’s frailty and the integration challenges that UBS faced. UBS has imposed several restrictions on Credit Suisse during the takeover, including lim- its on lending, spending, and contract sizes. These measures were seen as reasonable given the lapses in CS’s risk controls, although they could cause certain clients to leave the bank. On June 12, 2023, UBS successfully finalized its emergency acquisition of CS, thereby establishing a colossal Swiss banking institution with a balance sheet of $1.6 trillion and a robust foothold in wealth management. In tandem with this announce- ment, UBS revealed that CS will operate as a separate subsidiary. Additionally, CS’s bankers will be prohibited from acquiring new clients from high-risk countries or investing in complex financial products. These preventative measures, formulated by UBS’s com- pliance department, aimed to mitigate potential risks associated with the transaction. DATA AND METHODOLOGY High-frequency data We employ high-frequency data as the foundation of our single- event study. The use of high-frequency data offers two advantages over daily sampled data. First, it allows us to estimate the event’s impact more precisely on asset prices compared to low-frequency data. By narrowing the time window around the event, we can 36 See press release of Swiss National Bank of March 19, 2023: www.snb.ch/en/mmr/reference/ pre_20230319/source/pre_20230319.en.pdf minimize the influence of noise on the estimates, enabling us to accurately determine the event’s effect on asset prices.37 Second, high-frequency data enable us to account for differ- ent volatility regimes, which is crucial for statistical inference.38 This aspect is particularly important in our single-event analysis, as the statistical significance of abnormal returns is based on the time series of two single banks’ stock and bond returns, where we observe a significant increase in price volatility during the event compared to the estimation window.39 Although the relatively liquid CDS data allows us to analyze how the event impacted credit spreads intraday, it does not the detailed structure of the bond portfolio, including differences in duration and convexity across bonds. As the merging banks have multiple bond issues, each with its own return series, we rely on daily (end-of-the-day) bond prices to measure the market value of the bond portfolio accurately and precisely quantify the effects on bondholder wealth in USD. Data Intraday stock prices are sourced from Refinitiv Eikon while intra- day 5-year Credit Default Swap (CDS) data and daily bond prices and are collected from Bloomberg. The stock price and CDS spread data are sampled at 30-min intervals, and we limit the sample to observations from 9:30 to 17:30 (CET) to ensure a stock price and a CDS spread at each point in time. To facilitate the analysis, we convert the data into logarithmic stock returns and simple CDS changes. The observation at 9:30 represents the overnight stock log return or CDS change. In terms of the cross-section, our analysis covers CS, UBS, and a subset of six systemically significant European banks (BNP Paribas, Deutsche Bank, HSBC, Société Générale, Santander, and UniCredit).40 CS has a diverse range of bonds, with more than 90 in total. Among them, there are 59 corporate bonds, AT1 bonds, and 11 zero-coupon bonds that were traded in the bond markets. How- ever, we consider AT1 bonds as a separate category of capital that cannot be directly compared to the other bonds. Because they were fully written down and we do not further consider them in our event study. As for the zero-coupon bonds, Bloomberg does not report their prices, so we do not include them in our bond analysis. These bonds have a minimal impact on our bond analy- sis as their face value represents only a marginal portion (1.72%). Other studies have also excluded zero-coupon bonds due to their unique behavior and distinct characteristics compared to non- zero-coupon bonds.41 This leaves us with 59 corporate bonds for analysis. We remove two corporate bonds from the sample due to incomplete data or infrequent trading resulting in an insufficient number of observations for the analysis. Hence, our final bond sample consists of 57 corporate CS bonds with an approximate 37 Barclay, M. J., and R.H. Litzenberger. 1988. “Announcement Effects of New Equity Issues and the Use of Intraday Price Data.” Journal of Financial Economics 21: 71-99. 38 Elsas, R., and D.S. Schoch. 2023. “Robust Inference in Single Firm/Single Event Analyses.” Journal of Corporate Finance 80: 102391. 39 Fisch, J.E., J.B. Gelbach and J. Klick. 2018. “The Logic and Limits of Event Studies in Securities Fraud Litigation.” Texas Law Review 96: 553-621. 40 We constrain the sample to European banks due to the different trading hours in different geographical regions. We constrain the analysis to large banks due to the availability of CDS data. 41 Bessembinder et al. (2009). 17456622, 2025, 2, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/jacf.12674 by Fachhochschule N ordw estschw eiz, W iley O nline L ibrary on [15/01/2026]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense http://www.snb.ch/en/mmr/reference/pre_20230319/source/pre_20230319.en.pdf http://www.snb.ch/en/mmr/reference/pre_20230319/source/pre_20230319.en.pdf 109 face value of about 96.35 bn USD. This bond portfolio represents approximately 80% of CS’s total net long-term debt. Event study methodology We begin by estimating intraday abnormal stock log returns dur- ing the broad event window which spans from Friday, March 10, 2023, 17:30 to Friday, March 24, 2023, 17:30. This time frame provides us with 170 observations for abnormal returns and CDS changes. To measure abnormal returns, we employ factor models and estimate the necessary model parameters using data from the pre-event window which ranges from February 24 to March 10, comprising 180 observations. For stock log returns, our factor model incorporates the Stoxx Europe 600 index as a market factor and the Stoxx Europe 600 Banks index as an industry factor. Unfortunately, we do not have access to an intraday benchmark or estimating abnormal CDS changes. Instead, we construct a single factor using CDS changes from six systemically relevant large European banks, with equal weightings. However, the choice of the factor model is not expected to significantly impact the results of the event study. This is due to the substantial market movements observed for CS and UBS in comparison to other comparable stocks and CDSs. We define sub-event windows as follows: ∙ Weekend response (Day 0): The announcement of the CS takeover by UBS occurred on Sunday, March 19. Consequently, we define the most narrow event window (Day 0) as the sin- gle abnormal return from Friday, March 17, 17:30 to Monday, March 20, 09:30. ∙ Day (+1): The remaining trading day following the announce- ment is defined as event day (+1), spanning from Monday, March 20, 09:30 to Monday, March 20, 17:30. ∙ Days (+2; +3; +4; +5): The subsequent 24-h intervals are denoted as event days (+2) through (+5), encompassing Mon- day, March 20, 17:30 to Tuesday, March 21, 17:30, up to Thursday, March 23, 17:30 to Friday, March 24, 17:30. ∙ Days (−1; −2; −3; −4; −5): Similarly, the preceding 24-h inter- vals prior to the event are referred to as event days (−1) through (−5), covering Thursday, March 16, 17:30 to Friday, March 17, 17:30, up to Friday, March 10, 17:30 to Monday, March 13, 17:30. Cumulative event days combine the defined event time intervals above. For example, the event window Day (0; +1) corre- sponds to the time span between Day 0 plus Day 1, ranging from Friday, March 17, 17:30 to Monday, March 20, 17:30. Statistical inference We calculate two types of standard errors for statistical inference in our analysis. First, we follow the methods employed by Camp- bell, Lo, and MacKinlay in their 1997 book to calculate standard errors that consider the estimation of factor model parameters.42 42 Campbell, J., A. Lo, and A. MacKinlay. 1997, page 159, Equation 4.4.9. The Econometrics of Financial Markets, Princeton University Press. These standard errors incorporate the residual variance from the estimation window.43 We expect that the residual variance will be higher during the news-rich period within our event window. Since we employ high-frequency data, we can estimate the resid- ual variance in the event window using a comparable number of observations to the estimation window. Consequently, we present a modified version of standard errors that account for estima- tion error in the parameters and rely on the residual variance in the event window. Our approach aligns with the method- ology proposed by Ralf Elsas and Daniela Schoch, who have recently advocated the usage of high-frequency data in single event studies.44 RESULTS Stockholder wealth effects The results of our event study on cumulative abnormal stock returns (CAR) are presented in Figure 1 (30-min intervals) and Table 1 (sub-event windows). On the event weekend (Day 0), UBS experienced a significant abnormal stock return of −12%, while CS had a much larger abnormal return of −62%. Both returns were statistically significant based on t-statistics using either the estimation window residual variance (t-esw) or the event period residual variance (t-evw). However, both stocks showed recovery by the evening of the first trading day, with CARs of approximately 12% for UBS and CS on Day +1. By Day +2, CS stock had a CAR of 6%and UBS stock had a CAR of 8.9%. The two-day stockholder CAR (Day 0 to +2) was-55.16% for CS and +7.95% for UBS, corresponding to significant wealth effects. Both point estimates are economically large. In absolute terms, the CAR (0; +2) corresponds to stockholder wealth effects of +5.48 bn USD for UBS and -4.35bnUSD for CS. Given the large point estimate over the 2-day event window, we recognize that in the case of UBS, statistical significance at the 5% level is modest and imprecisely measured. The much larger t-statistics based on the residual variance of the estimation period (t-esw) suggest that the reason for low statistical significance comes from a much larger event window residual variance and underlines the importance of using high-frequency data. A closer examination of institutional equity ownership in CS (as a percentage of free float) based on Bloomberg data reveals some interesting trends. At the end of 2021, institutional 43 As Campbell, Lo, and MacKinlay show, it is common to exploit the cross-section of abnor- mal returns to take the event-specific variance into account. However, the cross-sectional approach is not feasible in our case because we have a single-event study. In such settings, Elsas and Schoch propose a parametric estimate of the residual variance based on intraday data in the event window. Due to the large market movements, particularly in the case of Credit Suisse, we choose a non-parametric estimate of the residual variance. For the same reason, were move the smallest 2.5% and largest 2.5% of the170 observations before estimating the residual variance in the event window. Elsas, R., and D.S. Schoch. 2023. “Robust Inference in Single Firm/Single Event Analyses.” Journal of Corporate Finance 80: 102391. 44 It is common to exploit the cross-section of abnormal returns to take the event-specific variance into account (Campbell et al., 1997). However, the cross-sectional approach is not feasible in our case because we have a single-event study. In such a setting, Elsas and Schoch (2023) propose a parametric estimate of the residual variance based on intraday data in the event window. Due to the large market movements, particularly in the case of Credit Suisse, we choose a non-parametric estimate of the residual variance. For the same reason, we remove the smallest 2.5% and largest 2.5% of the 170 observations before estimating the residual variance in the event window. 17456622, 2025, 2, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/jacf.12674 by Fachhochschule N ordw estschw eiz, W iley O nline L ibrary on [15/01/2026]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 110 JOURNAL OF APPLIED CORPORATE FINANCE F I G U R E 1 Event time figure: Cumulative abnormal stock returns. The figure reports the cumulative abnormal stock return (CAR) for Credit Suisse (blue) and UBS (red) from Friday, March 10, 2023, 17:30 to Friday, March 24, 2023, 17:30. Calculations are the same as described in the caption to Table 1. Confidence intervals are the point estimate ± two standard errors. The standard errors are based on the residual variance from the event window and account for an increased residual variance in the event window. ownership stood at 52.36%, increasing slightly to 59.23% by mid- 2022. However, just prior to the merger announcement event, there was a marked decrease in institutional ownership, drop- ping to 39.64%. Following the event, institutional ownership showed aconstanttrend:40.3% on the 26th of March (post-event), 39.03% on the 30th of April (post-event), and 38.31%by the end of May (post-event). We also investigate the short inter- est positions over the period from mid-2022 until February 2023, just prior to the event. We find that short positions have increased significantly from July 2022 (11,310,009 shares shorted) to December 2022 (28,584,988 shares shorted) and March 3023 (17,950,392 shares shorted). Overall, these patterns show that “smart” institutional investors offloaded their shares during the second half of 2022, suggesting that small retail investors suffered the most. Bondholder wealth effects We next analyze high-frequency data on corporate default swap (CDS) spreads. The cumulative abnormal CDS changes (CAC, expressed in basis points) are summarized in Figure 2 (30-min intervals) and Table 2 (sub-event windows as defined above). Over the event weekend they were −74 bp for CS and 7.6 bp for UBS. By the end of the first trading day (Day +1), the CDS changes were −4.5 bp for CS and +12.0 bp for UBS. On Tuesday (Day +2), CS saw a significant drop of −676 bp in CDS changes, while UBS had a smaller drop of 22.5 bp, both indicating significant changes. The 2-day CDS changes (Day 0 to +2) were −751 bp for CS (highly significant) and −3.0 bp for UBS (insignificant). These abnormal CDS spread changes suggest large positive abnormal returns for CS bondholders, although accurately measuring the bondholder wealth effects in USD faces challenges. First, the CDS is not necessarily representative of the value-weighted bond port- folio of CS. Second, large movements in the CDS that we observe are likely to render any approximation based on the yield-duration relationship inaccurate. To accurately measure the bondholder wealth effects in USD, we re-estimate at the daily frequency the bond return for a value- weighted CS bond portfolio. This portfolio has a face-value of about 96 bn USD (approximately 80% of CS’s total net long- term debt) and includes 57 bonds. Table 3 presents the results of cumulative abnormal daily bond returns for Credit Suisse (CS) during the event window from March 10, 2023, to March 24, 2023. The table provides results for CDS (Credit Default Swap) and two different averages: EW (the equally weighted bond portfolio average) and VW (the value-weighted bond portfolio average). The daily abnormal log bond returns are calculated using a one-factor model, with the factor being the iShares Core CHF Corporate Bond ETF (CH). The table displays the CAR per- centages, which represent the exponential function of the sum of the abnormal log returns over the specified event window minus one, expressed in percentage points. The “Day” column denotes the cumulative window in calendar days, as defined for the intraday data. The “t-esw” column provides the t-statistic for the CARs, based on the residual variance from the estimation window. We find a cumulative abnormal CS bond portfolio return of +34.74% (value-weighted) over the period from Day 0 to Day +2, which corresponds to an abnormal bondholder wealth increase of 22.65 bn USD. This abnormal USD bond return is computed as the market value of the bond portfolio containing 57 bonds on the evening of March 17, 2023, multiplied by the bondholder CAR (65.2 bn USD × +34.74% = 22.65 bn USD). Results, using the equally weighted portfolio are similar. 17456622, 2025, 2, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/jacf.12674 by Fachhochschule N ordw estschw eiz, W iley O nline L ibrary on [15/01/2026]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 111 TA B L E 1 Cumulative Abnormal Stock Returns. Credit Suisse UBS Day Time stamp CAR,% t-esw t-evw CAR,% t-esw t-evw −5 Fr,10/03,17:30: Mo,13/03,17:30 −3.82 −1.45 −0.52 −2.55 −2.72 −0.99 −4 Mo,13/03,17:30: Tu,14/03,17:30 −3.92 −1.51 −0.54 0.74 0.79 0.29 −3 Tu,14/03,17:30: We,15/03,17:30 −19.05 −7.80 −2.78 −3.24 −3.43 −1.25 −2 We,15/03,17:30: Th,16/03,17:30 17.91 6.23 2.22 1.84 1.95 0.71 −1 Th,16/03,17:30: Fr,17/03,17:30 −6.32 −2.48 −0.88 1.54 1.64 0.60 0 Fr,17/03,17:30: Mo,20/03,09:30 −62.05 −109.95 −39.16 −11.78 −40.13 −14.58 +1 Mo,20/03,09:30: Mo,20/03,17:30 11.70 4.21 1.50 12.11 12.28 4.46 +2 Mo,20/03,17:30: Tu,21/03,17:30 5.88 2.15 0.77 8.88 9.03 3.28 +3 Tu,21/03,17:30: We,22/03,17:30 −5.41 −2.11 −0.75 −4.39 −4.81 −1.75 +4 We,22/03,17:30: Th,23/03,17:30 −1.78 −0.68 −0.24 −1.70 −1.83 −0.66 +5 Th,23/03,17:30: Fr,24/03,17:30 −0.75 −0.28 −0.10 −0.43 −0.46 −0.17 0; +1 Fr,17/03,17:30: Mo,20/03,17:30 −57.65 −31.27 −11.14 −0.86 −0.89 −0.32 0; +2 Fr,17/03,17:30: Tu,21/03,17:30 −55.16 −20.05 −7.14 7.95 5.39 1.96 0; +5 Fr,17/03,17:30: Fr,24/03,17:30 −58.65 −12.79 −4.55 1.02 0.41 0.15 −5; +5 Fr,10/03,17:30: Fr,24/03,17:30 −65.80 −9.56 −3.41 −1.00 −0.25 −0.09 Note: This table reports cumulative abnormal stock returns for Credit Suisse and UBS during the event window from March 10, 2023, 17:30 to March 24, 2023, 17:30 (#170 return observations). Abnormal log returns are computed in 30 min intervals using a two-factor model. The factors are the Stoxx Europe 600 index and the Stoxx Europe 600 Banks index. Factor model parameters are estimated during an estimation window from February 24, 2023, 13:00 to March 10, 2023, 17:30 (#180 return observations). The cumulative abnormal return (CAR%) is the exponential function of the sum of the abnormal log returns over the specified event window minus one and is expressed in percentage points. “Day” reports the cumulation window in calendar days and, and “Time stamp” provides the precise cumulation window. t-esw is the t-statistic for the CARs as in Campbell, Lo, and MacKinlay (1997) and is based on the residual variance from the estimation window. t-evw is the t-statistic for the CARs based on the residual variance from the event window after truncating the residuals at the 2.5 and 97.5 percentiles and account for an increased residual variance in the event window. Additional notes Resolution of post event uncertainty We use the Day 0 to Day +2 CARs and CACs as our baseline results to assess the impact of the event on asset prices. The subsequent evolution of stock prices and CDS spread changes on Monday and Tuesday indicates that the weekend’s bailout-merger announcement was accompanied by announcement-related uncertainty, which was largely resolved between Monday and Tuesday afternoon. Two observations shed light on the vanishing announcement-related uncertainty within 2 days. Firstly, the CDS of CS exhibited a hesitant reaction at the Monday open and began to decline from Monday after 17:30 onwards (Day +2 in Table 2). Secondly, the CDS of UBS initially rose at the market open on Monday and continued to increase until the end of the day (Day 0; +1). However, this initial reaction proved to be temporary as it reversed on Tuesday (Day +2), result- ing in a 2-day change that was economically insignificant and lacked statistical significance. The behavior of UBS’s stock price mirrored the opposite patterns of the CDS from Monday’s open to Tuesday’s close, suggesting that the initial stock price response partly reflected announcement-related uncertainty. Pre-event response The release of the CS annual report on Tuesday (Day -4) had a minimal and insignificant impact on the CDS and equity returns. However, the announcement by Saudi National Bank to with- draw further financial support had a substantial impact on CS, reflected in a CAR of −19%, which was largely reversed on the following day. The CS CAC increased by 289 bp and contin- ued to rise on the subsequent day. In the case of UBS, we did not observe significant CARs, but we did observe a CAC of 15 basis points on Wednesday, which was statistically significant at the 10% level. There were no abnormal returns or abnormal CDS changes observed on Friday, March 17, for either CS or UBS, whether on a daily or intraday basis. Other European banks Table 4 presents the CARs and CACs for our control portfolio consisting of six major European banks, namely, BNP Paribas, Deutsche Bank, HSBC, Societe Generale, Santander, and Uni- Credit. Panel A shows the cumulative abnormal stock returns (CAR) for the six systematically relevant European banks. Panel B displays the cumulative abnormal CDS spread changes as a proxy for bondholder returns for the same set of banks. The mean values are provided for both panels. The data spans various time periods, indicated by the day and time stamps. In Panel A, the CAR percentages fluctuate across the banks and time periods. Notable movements include a decline in CAR just prior to the merger announcement, followed by a mix of positive and negative returns in subsequent days. The mean CAR for the six banks shows a negative value overall. Panel B focuses on the cumulative abnormal CDS spread changes, mea- 17456622, 2025, 2, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/jacf.12674 by Fachhochschule N ordw estschw eiz, W iley O nline L ibrary on [15/01/2026]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 112 JOURNAL OF APPLIED CORPORATE FINANCE F I G U R E 2 Event time figure: Cumulative abnormal CDS spread changes. The figure reports the cumulative abnormal CDS change (CAC) for Credit Suisse (blue) and UBS (red) from Friday, March 10, 2023, 17:30 to Friday, March 24, 2023, 17:30. Calculations are the same as described in the caption to Table 2. Confidence intervals are the point estimate ± two standard errors. The standard errors are based on the residual variance from the event window and account for an increased residual variance in the event window. TA B L E 2 Cumulative abnormal CDS spread changes. Credit Suisse UBS Day Time stamp CAC, bp t-esw t-evw CAC, bp t-esw t-evw −5 Fr, 10/03, 17:30: Mo, 13/03, 17:30 47.15 5.81 1.08 2.55 1.64 0.33 −4 Mo, 13/03, 17:30: Tu, 14/03, 17:30 53.79 7.16 1.34 7.41 5.14 1.03 −3 Tu, 14/03, 17:30: We, 15/03, 17:30 289.12 34.76 6.48 15.40 9.65 1.93 −2 We, 15/03, 17:30: Th, 16/03, 17:30 227.04 29.75 5.55 5.84 3.99 0.80 −1 Th, 16/03, 17:30: Fr, 17/03, 17:30 −80.58 −9.58 −1.79 11.55 7.16 1.43 0 Fr, 17/03, 17:30: Mo, 20/03, 09:30 −74.01 −36.19 −6.75 7.61 19.37 3.88 +1 Mo, 20/03, 09:30: Mo, 20/03, 17:30 −4.52 −0.62 −0.12 11.95 8.58 1.72 +2 Mo, 20/03, 17:30: Tu, 21/03, 17:30 −676.03 −87.13 −16.24 −22.51 −15.11 −3.03 +3 Tu, 21/03, 17:30: We, 22/03, 17:30 −44.61 −5.72 −1.07 −12.01 −8.02 −1.61 +4 We, 22/03, 17:30: Th, 23/03, 17:30 −25.14 −3.25 −0.61 0.20 0.13 0.03 +5 Th, 23/03, 17:30: Fr, 24/03, 17:30 −6.54 −0.78 −0.15 6.03 3.73 0.75 0; +1 Fr, 17/03, 17:30: Mo, 20/03, 17:30 −75.27 −9.62 −1.79 19.52 12.99 2.60 0; +2 Fr, 17/03, 17:30: Tu, 21/03, 17:30 −751.30 −66.71 −12.44 −2.99 −1.38 −0.28 0; +5 Fr, 17/03, 17:30: Fr, 24/03, 17:30 −827.59 −41.83 −7.80 −8.78 −2.31 −0.46 −5; +5 Fr, 10/03, 17:30: Fr, 24/03, 17:30 −294.33 −8.41 −1.57 34.03 5.06 1.01 Note: This table reports cumulative abnormal CDS changes for Credit Suisse and UBS during the event window from March 10, 2023, 17:30 to March 24, 2023, 17:30 (#170 CDS change observations). Abnormal CDS changes are computed in 30 min intervals using a one factor model. The factor is the equally weighted average CDS change of BNP Paribas, Deutsche Bank, HSBC, Societe Generale, Santander, and UniCredit. Factor model parameters are estimated during an estimation window from February 24, 2023, 13:00 to March 10, 2023, 17:30 (#180 CDS change observations). The cumulative abnormal CDS change (CAC) is the sum of the abnormal CDS change over the specified event window and is expressed in basis points. “Day” reports the cumulation window in calendar days and, and “Time stamp” provides the precise cumulation window. t-esw is the t-statistic for the CACDSs as in Campbell, Lo, and MacKinlay (1997) and is based on the residual variance from the estimation window. t-evw is the t-statistic for the CARs based on the residual variance from the event window after truncating the residuals at the 2.5 and 97.5 percentiles and account for an increased residual variance in the event window. 17456622, 2025, 2, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/jacf.12674 by Fachhochschule N ordw estschw eiz, W iley O nline L ibrary on [15/01/2026]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 113 TA B L E 3 Cumulative daily abnormal bond returns. CAR, % t-esw(CAR,%) Day Time stamp CDS EW VW CDS EW VW −5 Fr,10/03,Close: Mo,13/03,Close −0.49 −2.22 −2.51 −1.59 −4.94 −5.13 −4 Mo,13/03,Close: Tu,14/03,Close −0.98 −2.07 −2.30 −3.20 −4.64 −4.72 −3 Tu,14/03,Close: We,15/03,Close −26.62 −11.60 −12.99 −99.87 −27.17 −28.04 −2 We,15/03,Close: Th,16/03,Close 4.59 −2.87 −3.01 14.56 −6.46 −6.19 −1 Th,16/03,Close: Fr,17/03,Close −4.38 −1.95 −2.61 −14.49 −4.36 −5.34 0; +1 Fr,17/03,Close: Mo,20/03,Close 35.20 26.63 28.77 96.29 51.45 50.45 +2 Mo,20/03,Close: Tu,21/03,Close 5.14 4.34 4.63 16.14 9.33 9.11 +3 Tu,21/03,Close: We,22/03,Close 0.04 1.65 1.81 0.12 3.62 3.64 +4 We,22/03,Close: Th,23/03,Close −0.01 −0.89 −0.79 −0.02 −1.97 −1.60 +5 Th,23/03,Close: Fr,24/03,Close −3.22 −0.92 −1.24 −10.62 −2.05 −2.52 0; +2 Fr,17/03,Close: Tu,21/03,Close 42.15 32.12 34.74 78.50 42.44 41.58 0; +5 Fr,17/03,Close: Fr,24/03,Close 37.62 31.88 34.42 45.66 27.01 26.43 −5; +5 Fr,10/03,Close: Fr,24/03,Close −0.49 6.30 5.24 −0.49 4.17 3.19 Note: This table reports cumulative abnormal daily bond returns for Credit Suisse during the event window from March 10, 2023, to March 24, 2023, (#10 return observations). Below CDS are results for the CDS reference bond (ISIN: USH3698DCR38). EW refers to the equally-weighted average of 57 CS bonds. VW is the value-weighted average of 57 CS bonds (weighted by market value in USD). Daily abnormal log returns are computed using a one factor model, where the factor is the iShares Core CHF Corporate Bond ETF (CH). Factor model parameters are estimated during an estimation window from August 3, 2022, to March 10, 2023, (#152 return observations). The cumulative abnormal return (CAR,%) is the exponential function of the sum of the abnormal log returns over the specified event window minus one and is expressed in percentage points. “Day” reports the cumulation window in calendar days as defined for the intraday data. t-esw is the t-statistic for the CARs as in Campbell, Lo, and MacKinlay (1997) and is based on the residual variance from the estimation window. sured in basis points (bp), reflecting bondholder returns. Similar to Panel A, there is variation across the banks and time peri- ods. Some banks experienced significant increases or decreases in CDS spreads, indicating shifts in bondholder returns. The mean CAC (cumulative abnormal CDS spread changes) for the six banks also fluctuates throughout the observed periods. Over- all, we do not observe significant abnormal movements in equity or CDS returns during the event window for these banks. These findings suggest that the impact of the bailout-merger announce- ment event was confined to the two Swiss banks and that the performance and market reactions of the six other banks is not impacted by any changes in investor sentiment and expec- tations. This is significant as it underscores that the reaction in asset prices for the Swiss banks is directly linked to news about these two banks, rather than being a consequence of a broader shift in investor perception of risk or market-wide changes in discount rates. Consequently, our findings emphasize the unique impact of the merger-bailout event on the Swiss banking sector. Robustness Appendix Tables A.1 and A.2 present two robustness checks related to stockholder CARs. In Appendix Table A.1, we report stockholder CARs based on the constant mean return model. Fur- thermore, in Appendix Table A.2, we substitute the Banking factor with the equally weighted average of the six prominent European banks and calculate the stockholder CARs. Finally, in Appendix Table A.3, we replicate the abnormal CDS change results by replacing the factor model with a constant mean model. The results from all our robustness checks align closely with those of our baseline results mentioned earlier, indicating consistent findings across different models and methodologies. EXPLAINING THE WEALTH EFFECTS Bidding restrictions We argue that the observed stockholder wealth effects of the CS bailout-merger contradict what is typically observed in the litera- ture regarding competitive bids for failed banks. Firstly, the large positive acquirer stockholder wealth effect of 7.95% observed in this case deviates from prior research on bank mergers.45 Previ- ous studies extensively document negative abnormal returns for 45 Prior research shows that the bulk of positive takeover returns typically accrue to the stock- holders of the target firm and not the acquiring firm. For comprehensive reviews of empirical studies on short-run stock returns, see Alexandridis, G., N. Antypas, and N. Travlos. 2017. “Value Creation From M&As: New Evidence.” Journal of Corporate Finance 45: 632–650; Brooks, C., Z. Chen, and Y. Zeng. 2018. “Institutional Cross-Ownership and Corporate Strategy: The Case of Mergers and Acquisitions.” Journal of Corporate Finance 48: 187– 216; and Renneboog, L., and C. Vansteenkiste. 2019. “Failure and Success in Mergers and Acquisitions”, Journal of Corporate Finance 58: 650–699. Across different industries, acquirer announcement returns are often close to zero or indistinguishable from zero (see Netter, J., M. Stegemoller, and M.B. Wintoki. 2011. “Implications of Data Screens on Merger and Acquisi- tion Analysis: A large sample study of mergers and acquisitions from 1992 to 2009”, Review of Financial Studies 24, 2316–2357), and also low or zero for anticipated deals (see Tunyi, A.A., 2021. “Revisiting Acquirer Returns: Evidence From Unanticipated Deals”, Journal of Corpo- rate Finance 66, 101789). Some researchers report slightly positive acquirer announcement CARs. These include Asquith, P., R.F. Bruner, and D.W. Mullins Jr. 1983. “The Gains to Bidding Firms From Merger.” Journal of Financial Economics 11: 121–139; Eckbo, B. 1983. “Horizontal Mergers, Collusion, and Stockholder Wealth”, Journal of Financial Economics 11: 241–273; Martynova, M., and L. Renneboog. 2011. “The Performance of the European Mar- ket for Corporate Control: Evidence From the Fifth Takeover Wave”, European Financial Management 17: 208–259; and Alexandridis, G., N. Antypas, and N. Travlos. 2017. “Value 17456622, 2025, 2, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/jacf.12674 by Fachhochschule N ordw estschw eiz, W iley O nline L ibrary on [15/01/2026]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 114 JOURNAL OF APPLIED CORPORATE FINANCE TA B L E 4 Cumulative abnormal stock returns and CDS spread changes of systematically relevant European banks. Panel A: Cumulative abnormal stock returns, European banks CAR, % Day Time stamp BNP DB HSBC SG SANT UC Mean −5 Fr,10/03,17:30: Mo,13/03,17:30 0.36 3.64 3.14 0.91 −0.86 −3.45 0.62 −4 Mo,13/03,17:30: Tu,14/03,17:30 0.55 0.14 −0.92 −0.64 −1.11 0.71 −0.21 −3 Tu,14/03,17:30: We,15/03,17:30 −1.76 0.63 3.97 −3.87 1.14 −2.22 −0.35 −2 We,15/03,17:30: Th,16/03,17:30 0.51 −2.98 1.05 −2.16 0.26 0.83 −0.42 −1 Th,16/03,17:30: Fr,17/03,17:30 0.58 1.88 0.41 1.79 −2.27 −1.04 0.22 0 Fr,17/03,17:30: Mo,20/03,09:30 0.26 −2.24 1.40 −0.35 −0.43 0.19 −0.19 +1 Mo,20/03,09:30: Mo,20/03,17:30 1.71 −0.46 −2.41 −1.30 0.28 0.06 −0.35 +2 Mo,20/03,17:30: Tu,21/03,17:30 −0.28 0.35 −2.45 −0.93 −0.39 2.76 −0.16 +3 Tu,21/03,17:30: We,22/03,17:30 0.15 −1.68 1.83 −1.25 −0.27 −1.02 −0.37 +4 We,22/03,17:30: Th,23/03,17:30 1.29 0.26 0.95 1.15 0.44 0.41 0.75 +5 Th,23/03,17:30: Fr,24/03,17:30 −0.35 −3.05 1.94 −1.40 0.76 −1.11 −0.54 0; +1 Fr,17/03,17:30: Mo,20/03,17:30 2.35 −2.65 −1.15 −1.87 −0.20 0.02 −0.58 0; +2 Fr,17/03,17:30: Tu,21/03,17:30 2.06 −2.32 −3.57 −2.78 −0.59 2.78 −0.74 0; +5 Fr,17/03,17:30: Fr,24/03,17:30 3.16 −6.65 1.04 −4.26 0.34 1.02 −0.89 −5; +5 Fr,10/03,17:30: Fr,24/03,17:30 3.02 −3.66 9.04 −7.90 −2.47 −3.95 −0.99 Panel B: Cumulative abnormal CDS spread changes, European banks CAC, bp Day Time stamp BNP DB HSBC SG SANT UC Mean −5 Fr, 10/03, 17:30: Mo, 13/03, 17:30 2.55 3.61 4.33 3.38 5.99 0.05 3.32 −4 Mo, 13/03, 17:30: Tu, 14/03, 17:30 7.41 0.29 0.52 −0.96 −0.34 −1.28 0.94 −3 Tu, 14/03, 17:30: We, 15/03, 17:30 15.40 4.82 8.03 0.55 3.57 −1.51 5.14 −2 We, 15/03, 17:30: Th, 16/03, 17:30 5.84 1.15 13.62 −2.53 3.85 −0.49 3.57 −1 Th, 16/03, 17:30: Fr, 17/03, 17:30 11.55 −0.19 23.71 −6.20 8.58 3.38 6.81 0 Fr, 17/03, 17:30: Mo, 20/03, 09:30 7.61 3.36 −2.58 −1.68 3.73 2.68 2.19 +1 Mo, 20/03, 09:30: Mo, 20/03, 17:30 11.95 −3.22 7.11 −0.28 −5.09 −5.60 0.81 +2 Mo, 20/03, 17:30: Tu, 21/03, 17:30 −22.51 −6.28 −27.08 1.04 −6.75 2.29 −9.88 +3 Tu, 21/03, 17:30: We, 22/03, 17:30 −12.01 −4.71 11.59 −0.72 −6.02 −4.23 −2.68 +4 We, 22/03, 17:30: Th, 23/03, 17:30 0.20 0.49 13.37 −1.16 3.76 −2.16 2.42 +5 Th, 23/03, 17:30: Fr, 24/03, 17:30 6.03 4.10 24.18 −0.06 6.44 −5.81 5.81 0; +1 Fr, 17/03, 17:30: Mo, 20/03, 17:30 19.52 −0.22 7.73 −2.19 −1.69 −3.51 3.27 0; +2 Fr, 17/03, 17:30: Tu, 21/03, 17:30 −2.99 −6.50 −19.36 −1.15 −8.44 −1.22 −6.61 0; +5 Fr, 17/03, 17:30: Fr, 24/03, 17:30 −8.78 −6.62 29.79 −3.09 −4.26 −13.42 −1.06 −5; +5 Fr, 10/03, 17:30: Fr, 24/03, 17:30 34.03 3.41 76.81 −8.62 17.71 −12.67 18.44 Note: This table reports stockholder cumulative abnormal equity returns for six systematic European banks in Panel A, and bondholder cumulative abnormal returns, as proxied by abnormal CDS changes, in Panel B: BNP Paribas (BNP), Deutsche Bank (DB), HSBC Holdings, Société Générale (SG), Santander (SANT), UniCredit (UC). Mean refers to the mean CAR/CAC of the six individual banks. The calculations are the same as described in the caption to Tables 1 and 2. Creation From M&As: New Evidence.”, Journal of Corporate Finance 45: 632–650. While others, such as Morck, R., A. Shleifer, and R.W. Vishny. 1990, “Do Managerial Objectives Drive Bad Acquisitions?” Journal of Finance 45: 31—48; and Chang, S., and D.Y. Suk. 1998. “Failed Takeovers, Methods of Payment, and Bidder Returns.” Financial Review 33: 19–34; report slightly negative abnormal short-run announcement returns. Not even distressed target acquisitions typically create large acquirer abnormal returns. Rather, they are often negative (see Ang, J., and N. Mauck. 2011. “Fire Sale Acquisitions: Myth vs. Reality.” Journal of Bank- ing & Finance 35: 532–543) and even when positive, they average around 1% (see Oh, S. 2018. “Fire-Sale Acquisitions and Intra-Industry Contagion.” Journal of Corporate Finance 50, 265–293). 17456622, 2025, 2, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/jacf.12674 by Fachhochschule N ordw estschw eiz, W iley O nline L ibrary on [15/01/2026]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 115 acquiring banks in non-failed bank acquisitions.46 In the case of failed-bank acquisitions, acquirer returns are usually positive but significantly lower, ranging from 1% to 3%.47 Comparatively, the 2-day CAR of 7.95% for UBS stockholders is more than dou- ble the typical abnormal returns for bidders in Federal Deposit Insurance Corporation (FDIC) failed bank auctions in the US.48 It is likely that UBS stockholders benefited from the bidding restrictions imposed by SNB and Finma. Prior research indicates that the number of competing bidders in failed bank auctions pos- itively influences the bids submitted.49 This aligns with Richard Roll’s (winner’s curse hypothesis,50 suggesting that the winning bidder in a sealed-bid auction of an object with an uncertain value tends to overestimate its unobservable value. It appears that restricting the number of bidders to just one (UBS) was unnec- essary. Aside from BlackRock, which allegedly prepared a rival bid for CS, there were 23 global systemically important banks (G-SIBs) larger than UBS that could have participated in an auc- tion.51 Such an auction would likely have resulted in a higher transaction price.52 Consequently, the bidder restrictions proba- bly led to a transfer of wealth from CS to UBS shareholders, as indicated by the large negative abnormal stock return experienced by CS stockholders. While we are uncertain why no other banks were invited to sub- mit a bid for CS, prior research suggests that political access may have influenced this decision.53 Political access holds significant value for corporations. As Dennis Igan shows banks that engage in regulatory lobbying have a higher probability of winning auc- tions on more favorable terms.54 Theoretically, the industry-wide 46 See Bliss et al. (2007). “Learning by Observing: Information Spillovers in the Execution and Valuation of Commercial Bank M&As.” Journal of Finance 62: 181–216; as well as Houston et al. (2001), Sushka and Bendeck (1988). 47 Recently, Cowan and Salotti (2015) found that banks acquiring failed institutions (non- failed institutions) experience cumulative 2-day abnormal returns (CARs) in the magnitude of 3.23% (−0.05%). Their finding is in line with prior research, which finds 2-day CARs of 2.36% (see James, C.M., and P. Wier. 1987. “Returns to Acquirers and Competition in the Acquisition Market: The Case of Banking.” Journal of Political Economy 95: 355–370); 1.01% (Bertin et al, 1989) or 0.98% (Zhang, 1997); and a 3-day CAR of 1.16% (Cochran et al., 1995) or 1.94% (Baibirer et al., 1992). These studies investigate Federal Deposit Insurance Corporation (FDIC) failed bank acquisitions. The study of Baibirer et al. (1992) investigates Federal Home Loan Bank Board (FHLBB) supervised mergers of troubled savings and loan associations arranged by the Federal Savings and Loan Insurance Corporation. 48 Cowan and Salotti (2015). 49 Baibirer et al. (1992), Bertin et al. (1989), Cochran et al. (1995), Cowan and Salotti (2015), Gilberto and Varaiya (1989), and James and Wier (1987). 50 Roll (1986). 51 G-SIBs as of December 31, 2022, as measured by their balance sheet total. Data are from Bloomberg as per end of 2022, retrieved in April 2023. 52 A simple back-of-the-envelope calculation shows that the price attributed to the CS shares is at a very low level. Measured by Tobin’s q, a metric that is frequently used in the corporate finance literature, the value attributed to CS shares was the equivalent to 0.07 or one tenth of the average G-SIBs Tobin’s q of 0.73. The acquirer Tobin’s q amounted to 1.02. Using the price/sales-ratio (P/S), another frequently used valuation proxy, at the price attributed to CS shares, it amounts to 0.21, while the average P/S-ratio for G-SIBs (UBS) amounted to 1.67 (1.59). We use data from Bloomberg and values as per December 31, 2022. We employ Tobin’s q as a proxy for firms’ equity valuations and compute Tobin’s q as the market-value of equity to the book value of equity. It is essentially a market-to-book ratio for the entity. We compute the price-sales-ratio (P/S) as the market-value of equity scaled by net revenues. 53 Acemoglu, D., S. Johnson, A. Kermani, and J. Kwak. 2016. “The Value of Connections in Turbulent Times: Evidence From the United States,” Journal of Financial Economics 121: 368–391; Brown, J.R., and J. Huang. 2020. “All the President’s Friends: Political Access and Firm Value”, Journal of Financial Economics 138: 415–431; and Igan, D., T. Lambert, W. Wagner, and E.Q. Zhang. 2022. “Winning Connections? Special Interests and the Sale of Failed Banks.” Journal of Banking and Finance 140: 1—18. 54 Igan et al. (2022). lack of liquidity may have created a situation in which no other bidders were willing to make an offer for CS.55 The UBS/CS bailout-merger took place during a period of industry-wide shocks, with two US banks (Silicon Valley Bank and Signature Bank) being closed shortly before the event. This liquidity constraint would explain why the Swiss government, Finma, and SNB did not, to the best of our knowledge, develop any activity to receiving alternative offers within a reasonable timeframe. However, it seems that SNB and Finma should not have been caught off guard by the CS crisis. Bond spreads serve as warning signals for bank supervisors from the financial markets, typically rising as early as six quarters before a bank failure.56 This was also the case for CS, with numerous market prices indicating the approaching bank failure. For instance, the spread between AT1 bond prices of UBS and CS started to significantly widen from the first quarter of 2021, nine quarters before the merger-bailout, steadily increasing from approximately zero to over 20% by the end of the third quarter in 2022. Similarly, CS’s CDS spreads surpassed their 2007/2008 crisis levels in mid-2022, implying that Finma and SNB could have been prepared to resolve the bank or facilitate a merger involving multiple bidders if nec- essary. In addition, 6 months provided more than sufficient time to arrange a multi-bidder process for a bailout-merger with CS. Moreover, it is publicly known57 that top-level representatives of CS, UBS and the Swiss government engaged in active talks with respect to M&A, contingency planning and a potential merger of CS and UBS as early as December 2022. In the US, when a bank is on the verge of failing, FDIC typically allows only 90 days for the bank to take corrective actions, such as recapitalization or voluntary merger negotiations with a competitor. Concurrently, the FDIC begins structuring the resolution process. Considering international standards, there was ample time to organize a multi- bidder process for a CS bailout-merger. The Swiss government thus disposed of sufficient time to involve more than just one bidder into this merger-bailout. Coinsurance The positive bondholder wealth effects observed align with pre- vious research in principle. However, their magnitude appears unprecedented based on our knowledge. Matthew Billet58 and, more recently, Andriy Bodnaruk and Marco Rossi59 find a sig- nificant coinsurance effect, showing that target bonds exhibit significantly higher returns when the target’s rating is below that of the acquirer or when the combination is anticipated to reduce 55 See Shleifer, A., and R.W. Vishny (1992). “Liquidation Values and Debt Capacity: A Market Equilibrium Approach.” Journal of Finance 47: 1343–1366; and Acharya, V., and T. Yorul- mazer. 2007. “Cash-In-The-Market Pricing and Optimal Resolution of Bank Failures.” Review of Financial Studies 21: 2705–2742. 56 See Jagtiani, J., and C. Lemieux. 2001. “Market Discipline Prior to Bank Failure.” Journal of Economics and Business 53: 313–324. 57 See page 39 of UBS Group AG filing with the Securities Exchange Commission (SEC) as of May 16, 2023: Amendment No. 1 to Form F-4, registration statement under Securities Act of 1933, registration no. 333–271453 58 Billet et al. (2004). 59 Bodnaruk and Rossi (2016) 17456622, 2025, 2, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/jacf.12674 by Fachhochschule N ordw estschw eiz, W iley O nline L ibrary on [15/01/2026]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 116 JOURNAL OF APPLIED CORPORATE FINANCE target risk.60 This holds true for the UBS/CS merger. Sheng Syan Chen et al.61 also support the notion that coinsurance can sig- nificantly impact bondholder abnormal returns. Nonetheless, the effect of coinsurance is likely to be much smaller than what we observe in the present case. For example, Matthew Billet et al. find an average positive mean excess return of 4.30% for below investment-grade target bonds.62 The 34.74% or 22.65 bn USD bondholder wealth increase that we find thus appears to be extraordinarily high. The AT-1 bond write-down may explain some portion of it: Risky debt may, in several states of the world, not be paid in full. This applies specif- ically to banks, as they are highly levered by construction. Any intervention that increases equity or equity like capital provides a safety cushion to debt and should thus increase its value. Writ- ing down AT-1 bonds may thus represent the equivalent to an equity injection, resulting in coinsurance. Veronesi and Zingales, who investigate the US government interventions in the finan- cial sector announced in 2008, find a coinsurance effect in the magnitude of 29% of the value of new money (equity) invested by the government, which accrued to bondholders.63 Immedi- ately prior to the event, on the evening of March 17, 2023, the equally weighted portfolio of CS’s AT-1 bonds traded at an approximate value of 22.9% of face value or an approximate 3.9 bn USD. Even if all of the AT-1 write-down created a coinsurance effect, we are still left with a high unexplained bondholder wealth increase. One other way to look at coinsurance is to investigate the asset quality of the two banks prior to the bailout-merger agreement. In Table 4 we provide various metrics related to asset quality and capital adequacy for the two banks. The data includes historical averages, standard deviations, and other relevant indicators over different time periods. Under the “Asset Quality (%)” section, metrics such as loan loss coverage, loan loss provision/total loans, loan loss reserves/total loans, net charge-offs/average loans, and other similar ratios are presented. These ratios measure the banks’ ability to handle loan losses and their loan loss reserve coverage. The “Capital Adequacy” section focuses on metrics related to cap- ital strength and risk-based capital ratios. Common equity tier 1 capital ratio, tier 1 risk-based capital ratio, leverage ratio, and total capital adequacy ratio are included. These indicators assess the banks’ capital adequacy and their ability to absorb losses and manage risks. These metrics are crucial in evaluating the banks’ ability to manage risks, maintain sufficient capital, and safeguard against potential financial stress. Overall, the descriptive statistics in the table suggest that the asset quality metrics of CS are generally weaker compared to those of UBS. This indicates that CS bondholders are more likely to gain from a merger that reduces overall asset risk compared to UBS bondholders, which can be attributed to the coinsurance effect. The coinsurance effect implies that a merger can result in a more 60 This is the case here. The acquirer’s ratings are significantly higher than those of the target: As per the event date, the acquirer (UBS) disposed of a bank operating companies Moody’s (Fitch, S&P) Aa3 (AA-, A1) rating, while the target was ranked some two notches lower at A3 (BBB+, A-). Moreover, the target’s rankings of Moody’s and Fitch were on a negative outlook. Rating information is from the CS fixed income investor presentation dated March 14, 2023. 61 Chen et al. (2022). 62 Billet et al. (2004). 63 Veronesi and Zingales (2010). diversified and robust asset base, reducing the risk for bondholders and potentially benefiting CS bondholders more significantly in this context. Too-big-to-fail and taxpayer’s cost The “too big to fail” (TBTF) effect refers to the perception that certain financial institutions, particularly large banks, are deemed so crucial to the functioning of the economy that their fail- ure could have severe systemic consequences. As a result, these banks may receive special treatment or support from the gov- ernment and regulators to prevent their collapse. Marıa Fabiana Penas and Haluk Unal demonstrate that achieving TBTF status leads to abnormal returns for bondholders in merger situations.64 The bailout-merger between UBS and CS has resulted in the creation of a new bank that undeniably falls under the TBTF cat- egory. Therefore, it is highly unlikely that CS bondholders will experience any defaults in the near future. Our findings suggest that bond markets strongly believe in TBTF policies. Quantifying which effect (coinsurance or TBTF) dominates is, however, chal- lenging. We contend that the combination of coinsurance and the TBTF effect largely explains the observed wealth effects for CS bondholders. It is not feasible to estimate the total economic costs of the TBTF effect. However, we can assess it indirectly by looking at changes in the financing costs for the Swiss government. The government has certainly taken some risk to rescue CS, and its sovereign credit risk may have been affected by doing so. Acharya et al. (2014) describe the theoretical and empirical relationship between sovereign credit risk and bank bailouts. They evidence a loop between the financial sector and sovereign credit risk and present evidence that bank bailouts transfer risk from bank bal- ance sheets to sovereigns, triggering the rise in sovereign credit risk during and after the global financial crisis (GFC). Using the same methods as Viral Acharya et al, in their 2014 Journal of Finance article,65 we use CDS spreads to assess whether Switzerland’s sovereign credit risk has increased due to the merger-bailout. Figure 3 shows the CDS spreads of Switzerland throughout the event period, alongside several European coun- tries. CDS spreads of Switzerland moved in line with that of other European countries until the merger-bailout. Germany, the Netherlands, and Sweden show relatively stable CDS spreads dur- ing the event window, while Switzerland’s CDS spread (depicted by the red solid line) experiences a significant increase. The CDS spreads of UK and France also respond to the event, although their CDS spread changes are more gradual. Before the event, Switzer- land’s CDS spread hovers around 11 basis points, lower than Germany, the Netherlands, and Sweden, which range from 13 to 15 basis points. However, on the Friday just prior to the event, Switzerland’s CDS spread jumps from 11.16 basis points to 20.08 basis points and maintains this level until the following Friday of the first trading week after the bailout-merger. Subsequently, it further increases to 25.01 basis points on the subsequent 64 Penas and Unal (2004). 65 Acharya, V., J. Drechsler, and P. Schnabl. 2014, “A Pyrrhic Victory? Bank Bailouts and Sovereign Credit Risk.” Journal of Finance 69: 2689–2739 17456622, 2025, 2, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/jacf.12674 by Fachhochschule N ordw estschw eiz, W iley O nline L ibrary on [15/01/2026]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 117 F I G U R E 3 Sovereign CDS spreads: Before and after the credit Suisse—Bailout. The figure reports the CDS spreads for Switzerland (red solid line), Germany (black solid line), Sweden (grey dotted line), the Netherlands (grey dots), the United Kingdom (blue solid line), and France (blue dots). Data are from Bloomberg. We use daily mid-prices. Monday. Throughout the 60-day period following the event, Switzerland’s CDS spread remains at an average level of 20.4 basis points. Sven Klingler and David Lando show that there may be a disconnect between changes in bond yield spreads and CDS pre- miums.66 We therefore additionally estimate CARs on an equally weighted portfolio of twenty-two Swiss government bonds. The result is shown in Table 5. The yield of Swiss government bonds increased by a significant 16 basis points from Friday to Tuesday close, approximately matching the observed increase in the CDS spreads (see Figure 3). Our findings suggest that the merger-bailout has had a dis- cernible impact on Switzerland’s sovereign credit risk, as reflected by the significant jump in CDS spreads, reflected by an abnor- mal approximate increase in Switzerland’s sovereign bond yield in the amount of 16 basis points. The elevated and sustained levels of the CDS spread indicate increased market perceptions of risk associated with Switzerland’s creditworthiness during and after the event. Lending from prior research, an increase in Switzerland’s credit risk should affect Switzerland’s expected cost of debt.67 To gauge for this effect in USD, we estimate the expected present value of Switzerland’s increased cost of debt. We assume that each matur- ing bond is immediately replaced at the due date by the issuance 66 Klingler, S., and D. Lando (2018). “Safe Haven CDS Premiums.” Review of Financial Studies 31: 1856—1895. 67 Elton, E.J., M.J. Gruber, D. Agrawal, and C. Mann. 2001. “Explaining the Rate Spread on Corporate Bonds.” Journal of Finance 56: 247–277; Collin-Dufresne, P., R.S. Goldstein, and J.S. Martin. 2001. “The Determinants of Credit Spread Changes.” Journal of Finance 56: 2177–2207; and Campbell, J.Y., and G.B. Taksler. 2003. “Equity Volatility and Corporate Bond Yields.” Journal of Finance 58: 2321—2349. TA B L E 5 Cumulative abnormal yield change of Swiss government bonds. Swiss government bonds (EW) Day Time stamp CAC, bp t-esw −5 Fr, 10/03, close: Mo, 13/03, close −21.73 −3.74 −4 Mo, 13/03, close: Tu, 14/03, close 10.28 1.77 −3 Tu, 14/03, close: We, 15/03, close −24.82 −4.27 −2 We, 15/03, close: Th, 16/03, close 2.54 0.44 −1 Th, 16/03, close: Fr, 17/03, close −9.52 −1.64 0;+1 Fr, 17/03, close: Mo, 20/03, close 0.68 0.12 +2 Mo, 20/03, close: Tu, 21/03, close 15.17 2.61 +3 Tu, 21/03, close: We, 22/03, close 7.08 1.22 +4 We, 22/03, close: Th, 23/03, close −4.96 −0.85 +5 Th, 23/03, close: Fr, 24/03, close −3.70 −0.64 0; +2 Fr, 17/03, close: Tu, 21/03, close 15.85 1.92 0; +5 Fr, 17/03, close: Fr, 24/03, close 14.26 1.09 −5; +5 Fr, 10/03, close: Fr, 24/03, close −28.99 −1.54 Note: Cumulative abnormal yield changes (CAC) are calculated for an equally weighted port- folio of all outstanding Swiss government bonds with available trading data (see Table 6). The data are sampled at the daily frequency (close-to-close). t-esw is the t-statistic as in Campbell, Lo, and MacKinlay (1997) and is based on the residual variance from the estimation window. of a new bond of the same nominal amount. For each sovereign bond in the portfolio, we calculate and then discount the marginal increase in the cost of debt of between 8 and 10 bp with the discount factor (1+YTM)t back to June 1, 2023, whereas YTM 17456622, 2025, 2, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/jacf.12674 by Fachhochschule N ordw estschw eiz, W iley O nline L ibrary on [15/01/2026]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 118 JOURNAL OF APPLIED CORPORATE FINANCE TA B L E 6 Switzerland’s marginal increase in the cost of debt. (I) 10 bp increase (II) 8 bp increase ISIN Issue amount CHF Term years YTM ΔCD PV ΔCD PV CH0127181177 3,343.40 1.04 0.0111 3.34 3.30 2.67 2.64 CH0184249990 3,760.65 2.18 0.0100 3.76 3.67 3.01 2.94 CH0224396983 3,803.48 3.03 0.0095 3.80 3.69 3.04 2.95 CH0031835561 3,080.52 4.13 0.0092 3.08 2.96 2.46 2.37 CH0008680370 5,612.46 4.93 0.0090 5.61 5.37 4.49 4.29 CH0224397346 4,613.03 6.15 0.0090 4.61 4.37 3.69 3.49 CH0224397171 3,500.09 7.09 0.0090 3.50 3.29 2.80 2.63 CH0127181029 3,378.47 8.18 0.0090 3.38 3.14 2.70 2.51 CH0344958688 3,111.42 9.21 0.0092 3.11 2.86 2.49 2.29 CH0015803239 4,555.96 10.00 0.0095 4.56 4.16 3.64 3.33 CH0440081393 2,272.17 11.23 0.0098 2.27 2.04 1.82 1.63 CH0557778310 2,165.78 12.24 0.0100 2.17 1.92 1.73 1.54 CH0024524966 4,305.89 12.96 0.0102 4.31 3.79 3.44 3.03 CH0127181193 4,054.08 14.28 0.0103 4.05 3.51 3.24 2.81 CH0440081567 1,412.15 15.63 0.0103 1.41 1.20 1.13 0.96 CH0440081401 2,532.25 16.38 0.0102 2.53 2.14 2.03 1.71 CH0127181169 4,365.97 19.19 0.0102 4.37 3.59 3.49 2.87 CH0344958498 3,650.15 22.40 0.0101 3.65 2.91 2.92 2.33 CH0009755197 2,790.82 25.98 0.0095 2.79 2.15 2.23 1.72 CH0344958472 2,517.72 32.44 0.0091 2.52 1.85 2.01 1.48 CH0224397338 2,186.35 35.51 0.0083 2.19 1.58 1.75 1.27 CH0224397007 3,408.21 41.67 0.0132 3.41 2.41 2.73 1.93 Issue amount weighted YTM: 0.0098 ∑PV = 65.90 ∑PV = 52.72 Present value of taxpayer’s cost: CHF 6,737.59 CHF 5,390.07 (=∑PV/ weighted YTM) USD 7,209.22 USD 5,767.36 Note: The table reports for the outstanding Swiss government bonds the issue amount, the remaining term in years, and the yield to maturity (YTM). Scenario (I) reports the change in cost of dept assuming that refinancing costs are persistently 10 bp higher (ΔCD). PV is the present-value of the respective increase in the cost of dept assuming that the higher refinancing costs become effective at the due date and a complete rollover of the dept. Scenario (II) reports the same calculations assuming a persistent 8 bp increase. The present value of taxpayer’s cost is calculated as the perpetuity value of the sum of the present-values of the increase in cost of debt using the issue amount weighted YTM. The data are from Bloomberg as of June 12, 2023. is the yield to maturity of the respective bond, and t is the remaining term of the respective bond in years. The present value of the expected marginal increase in the cost of debt amounts to approximately 56.0–70.5 million USD, respectively. Next, we capitalize this expected increase in the cost of debt. To this end, we use Switzerland’s existing bond portfolio, its value weighted remaining term (t) and the value weighted dis- count factor (1+YTM)t to approximate the present value by assuming a perpetuity (see the details in Table 6). Based on our estimation, the capitalized value of the expected jump in Switzerland’s cost of debt amounts to an approximate 5.8–7.2 bn USD.68 68 We acknowledge that these amounts are gross of any fees and charges that shall be paid by UBS/CS to the government. Such fees shall obviously reduce the present value of the increase in Switzerland’s cost of debt. They are, however, expected to be of short-term nature only and therefore of relatively small magnitude. CONCLUSIONS We show that the UBS-CS-merger substantially impacted the wealth of the participating firms’ stockholders and bondholders. It created a net value of 19.5 bn USD, distributed to UBS stock- holders (5.1 bn USD), CS stockholders (−4.4 bn USD), and CS bondholders (18.8 bn USD). The combined wealth effect cannot be explained by the participating firms’ abnormal returns on secu- rities. While the Swiss government claims that the bailout-merger is a private transaction that has the potential to come at zero cost to the taxpayer, we find that there have likely been large transfers of wealth from taxpayers to UBS/CS stakeholders. We identify various channels that may have created this sur- prisingly large, combined wealth effect. First, we argue that UBS stockholders have profited from bidding restrictions imposed by the government. These bidding restrictions may be the result of political ties between the government and top-level representatives of UBS and CS, who engaged in meetings to discuss the potential 17456622, 2025, 2, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/jacf.12674 by Fachhochschule N ordw estschw eiz, W iley O nline L ibrary on [15/01/2026]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 119 merger and other contingency plans as early as in December 2022. Second, we believe that CS bondholders profited from substantial coinsurance effects. Third, the “too-big-to-fail” channel, com- bined with a material loss protection agreement which covered a specific portfolio of CS assets (corresponding to approximately 3% of the combined assets of the merged bank) may have con- tributed to the combined wealth effect. Finally, and importantly, we infer from our analysis that the government intervention likely came at the cost of a significant jump in Switzerland’s sovereign credit risk and thus an increase in its expected cost of debt, implying the risk of a substantial taxpayer wealth transfer in the magnitude of approximately six to seven bn USD. It seems that the reforms adopted after the 2007–2009 crisis still fall short in resolving issues with systemically important bank institutions. Staggering costs of extensive government interven- tion in a banking crisis, as described by Veronesi and Zingales for the US financial sector during the 2008 global financial cri- sis, seem to be inherent in the banking system.70 As in the GFC, and described in more detail by Anjan Thakor in 2015,71 taxpay- ers and private investors still appear to bear the bailout costs for failing banks. Authorities act late, apply corrections only after the risks of failure have become severe. Both the failure of bank exec- utives and the deficit of supervisors to anticipate necessary tasks in case of an intervention (such as avoiding unnecessary restrictions on bidder participation) have created costly inefficiencies in the bailout process, including substantial wealth transfers from tax- payers to the banking sector. Restoring confidence to the financial system should have been achieved at a lower cost. K E Y WO R D S additional tier 1 (AT1), bank bailouts, banking crises, contingent convertible bonds (CoCo), credit risk subsidies, emerging market crises, fire-sale, macro- governance failures, regulation, systemically important banks (SIB), too-big-to- fail R E F E R E N C E S Acemoglu, D., S. Johnson, A. Kermani, and J. Kwak. 2016. “The Value of Con- nections in Turbulent Times: Evidence From the United States.” Journal of Financial Economics 121: 368–91. Alexandridis, G., N. Antypas, and N. Travlos. 2017. “Value Creation From M&As: New Evidence.” Journal of Corporate Finance 45: 632–50. Acharya, V., and T. Yorulmazer. 2007. “Cash-in-the-Market Pricing and Optimal Resolution of Bank Failures.” Review of Financial Studies 21: 2705–42. Acharya, V., I. Drechsler, and P. Schnabl. 2014. “A Pyrrhic Victory? Bank Bailouts and Sovereign Credit Risk.” Journal of Finance 69: 2689–739. Amiram, D., Z. Bozanic, J.D. Cox, Q. Dupont, J.M. Karpoff, and R. Sloan. 2018. “Financial Reporting Fraud and Other Forms of Misconduct: A Mul- tidisciplinary Review of the Literature.” Review of Accounting Studies 23: 732–83. Ang, J., and N. Mauck. 2011. “Fire Sale Acquisitions: Myth vs. Reality.” Journal of Banking & Finance 35: 532–43. Asquith, P., R.F. Bruner, and D.W. Mullins Jr. 1983. “The Gains to Bidding Firms From Merger.” Journal of Financial Economics 11: 121–39. Atanasov, V., V. Ivanov, and K. Litvak. 2012. “Does Reputation Limit Oppor- tunistic Behavior in the VC Industry? Evidence From Litigation Against VCs.” Journal of Finance 62: 2215–46. Baibirer, S. D., G.D. Jud, and F.W.