Managerial risktaking behavior in both financial and nonfinancial firms has been an attractive focus for the lens of many researchers (Hubbard and Palia, 1995; Houston and James, 1995; Knopf, Nam and Thornton, 2002; Coles, Daniel, and Naveen, 2006; Chen, Steiner and Whyte, 2006; Acharya and Naqvi, 2012). Excessive CEO risktaking in the financial sector especially has been blamed for playing a crucial role in the build up to the 20082009 financial crisis. Acharya and Naqvi (2012) develop a theoretical model to show that bank overlending may result from managers’ desire to receive higher compensation in the presence of an agency problem between a bank manager and shareholders.1 Other studies have revealed a positive correlation between option compensation and risktaking incentives, thus increasing bank risk taking and bankspecific default risk (Jeitschko and Jeung, 2005; Mehran and Rosenberg, 2007; Balachandran, Kogut and Harnal, 2010; Bebchuk, Cohen and Spamann, 2010; Fahlenbrach and Stulz, 2011; Hagendorff and Vallascas, 2011). For example, Coles, Daniel and Naveen (2006) suggest that the higher Vega gives executives incentive to implement more aggressive debt policy and invest more in riskier assets (e.g. RD). Similarly, DeYoung, Peng, and Yan (2013) show that banks in which CEOs have high risktaking incentives (highVega banks) exhibit substantially larger amounts of both systematic and idiosyncratic risk.2 To some extent, risktaking is good and that is why CEOs are given ESOPs (employee stock ownership plans) and equity stake to converge their interest with those of the shareholders. The problem is when CEOs go overboard and take “excessive” risk which is higher than the optimal level. Although above studies have confirmed that CEO risktaking
元 智 大 學 管理學院商學博士班 (財務金融學程) 博 士 論文 銀行經理人風險承擔動機的黑暗面：從銀行放款決策分析之 The Dark Side of Bank CEO Risk-taking Incentives: Evidence from Bank Lending Decisions 研 究 生： 指導教授： 陳氏垂玲 駱建陵 林智勇 中華民國 一百零九 年 六月 銀行經理人風險承擔動機的黑暗面：從銀行放款決策分析之 The Dark Side of Bank CEO Risk-taking Incentives: Evidence from Bank Lending Decisions 研 究 生: 陳氏垂玲 Student: TRAN THI THUY LINH 指 導 教 授: 駱建陵 Advisors: Prof CHIEN-LING LO 林智勇 Prof CHIH-YUNG LIN 元智大學 管理學院博士班（財務金融學程） 博士論文 A Dissertation Submitted to Doctor of Philosophy Program College of Management Yuan Ze University in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy in Finance June 2020 Chungli, Taiwan, Republic of China 中華民國 一百零九 年 六月 ii 銀行經理人風險承擔動機的黑暗面：從銀行放款 決策分析之 研究生: 陳氏垂玲 指導教授: 林智勇 駱建陵 元智大學 管理學院商學博士班 (財務金融學程) 摘要 本研究探討銀行首席執行官的冒險動機（vega）如何影響銀行貸款決策 該研究 的實證結果表明 vega 與貸款公告周圍的累積異常收益（CARs）顯著負相關, 這 證明了 vega 對銀行貸款市場具有真正的影響 另外, 根據現有的 CEO 激勵文獻, 我們發現具有較高冒險精神的 CEO 傾向於放寬銀行貸款合約中的貸款標準，以 尋求更高的報酬 有證據表明, 維加係數較高的銀行傾向於收取較低的貸款利差, 要求較少的貸款契約, 並且尋求抵押品的可能性較低 結果將會變弱, 這可支持以 下觀點：高 CEO 冒險行為可能會在銀行經理和股東之間造成代理問題 經濟文學雜誌: G21, G32, G34 關鍵字：CEO激勵, 銀行貸款合約, 累積超額收益, 公司治理, 代理問題 iii The Dark Side of Bank CEO Risk-taking Incentives: Evidence from Bank Lending Decisions Student: Tran Thi Thuy Linh Advisors: Chien-Ling Lo Chih-Yung Lin Doctor of Philosophy Program Major of Finance College of Management Yuan Ze University Abstract This paper investigates how bank CEO risk-taking incentives(vega) influence bank lending decisions Empirical finding of the study reveals that vega is significantly negatively related to cumulative abnormal returns(CAR) around loan announcements, confirming that vega has a real effect on the bank loan market In addition, consistent with the existing CEO incentive literature, we find that CEOs with higher risk-taking incentives tend to relax their lending standards in bank loan contracts to pursue higher compensation Evidence shows that banks with high vega tend to charge a significantly lower loan spread, demand fewer loan covenants, and have lower probability to seek collateral Results become weaker when banks have strong corporate governance mechanisms, supporting the proposition that high CEO risk-taking incentives may create an agency problem between a bank manager and shareholders JEL: G21, G32, G34 Keywords: CEO incentives, bank loan contracts, cumulative abnormal returns, corporate governance, agency problem iv Acknowledgements First and foremost, I would like to express my deep and sincere gratitude to my advisor, Professor Chih-Yung Lin for giving me the opportunity to research and providing invaluable guidance throughout this research His dynamism, vision, sincerity and motivation have deeply inspired me He has taught me the methodology to carry out the research and to present the research works as clearly as possible It was a great privilege and honor to work and study under his guidance Without his persistent help, the goal of my dissertation would not have been realized I am extremely grateful for what he has offered me I would also like to express my sincere thanks to Professor Chien-Ling Lo and Professor Po-Hsin Ho, who support and help me a lot in the journey towards this degree My appreciation also extends to all Professors at College of Management who have given me a great deal of knowledge in the last four years My thanks go to all my classmates and friends in Yuan Ze university for their valuable shares Last but not least, I would like to show my endless love to my parents, my husband and my son Thank for all coming to my life, always accompanying and staying by my side v Table of Contents 摘要 iii Abstract iv Acknowledgements v Table of Contents vi List of Tables vii Chapter Introduction Chapter Hypothesis development Chapter Data and methodology 11 3.1 Data and other variables 11 3.2 Summary Statistics 12 3.3 Methodology 14 Chapter Empirical Results 17 4.1 Bank’s CEO risk-taking incentives and cumulative abnormal returns(CARs) 17 4.2 Bank’s CEO risk-taking incentives and bank loan spread 17 4.3 Bank’s CEO risk-taking incentives and non-price terms 19 4.4 Bank’s CEO risk taking behavior: A Difference-in-Difference analysis 19 4.5 CEO risk-taking-incentive effect: Corporate-governance channels 21 4.6 Robustness checks 24 4.6.1 Control for CEO characteristics and other compensation schemes 24 4.6.2 Bank size analysis 24 4.6.3 Change regression 25 4.6.4 Bank’s CEO risk-taking incentives and cumulative abnormal returns (CARs) 26 Chapter Conclusion 28 References 29 Appendix A: Variable definition 52 Appendix B: Sample banks 54 Appendix C: Vega measure 55 vi List of Tables Table Descriptive Statistics 34 Table Loan, borrower, and lender characteristics for banks with high and low Vega 35 Table Correlation matrix 36 Table Bank’s CEO risk-taking incentives and cumulative abnormal returns (CARs) 37 Table Bank’s CEO risk-taking incentives and bank loan spread 39 Table Bank’s CEO risk-taking incentives: Non-price terms 41 Table CEO risk-taking incentives effect: Difference-in-Difference analysis 43 Table CEO risk-taking incentives and bank loan spread: Evidence from bank governance channels 45 Table Robustness check (I): Control for CEO characteristics and other CEO compensation 46 Table 10 Robustness check (II): Bank size analysis and the change regression 48 Table 11 Robustness check(III): Bank’s CEO risk-taking incentives and cumulative abnormal returns 50 vii Chapter Introduction Managerial risk-taking behavior in both financial and non-financial firms has been an attractive focus for the lens of many researchers (Hubbard and Palia, 1995; Houston and James, 1995; Knopf, Nam and Thornton, 2002; Coles, Daniel, and Naveen, 2006; Chen, Steiner and Whyte, 2006; Acharya and Naqvi, 2012) Excessive CEO risk-taking in the financial sector especially has been blamed for playing a crucial role in the build up to the 2008-2009 financial crisis Acharya and Naqvi (2012) develop a theoretical model to show that bank over-lending may result from managers’ desire to receive higher compensation in the presence of an agency problem between a bank manager and shareholders.1 Other studies have revealed a positive correlation between option compensation and risktaking incentives, thus increasing bank risk taking and bank-specific default risk (Jeitschko and Jeung, 2005; Mehran and Rosenberg, 2007; Balachandran, Kogut and Harnal, 2010; Bebchuk, Cohen and Spamann, 2010; Fahlenbrach and Stulz, 2011; Hagendorff and Vallascas, 2011) For example, Coles, Daniel and Naveen (2006) suggest that the higher Vega gives executives incentive to implement more aggressive debt policy and invest more in riskier assets (e.g R&D) Similarly, DeYoung, Peng, and Yan (2013) show that banks in which CEOs have high risk-taking incentives (high-Vega banks) exhibit substantially larger amounts of both systematic and idiosyncratic risk.2 To some extent, risk-taking is good and that is why CEOs are given ESOPs (employee stock ownership plans) and equity stake to converge their interest with those of the shareholders The problem is when CEOs go overboard and take “excessive” risk which is higher than the optimal level Although above studies have confirmed that CEO risk-taking Acharya and Naqvi (2016) show that, if a bank is awash with deposits from investors, its manager will be more likely to undertake high-risk projects to pursue his/her own self-interest and to sanction excessive loans by lowering lending rates and loosening lending standards (underprice the risk of projects), leading to asset-price bubbles and sowing seeds of future bank failure Gande and Kalpathy (2017) indicate that equity incentives (Vega) embedded in CEO compensation contracts are positively associated with risk taking in financial firms and result in potential solvency problems incentives increase bank risk exposure, how such exposure affects bank lending decisions has not to date been examined Specifically, in this paper we investigate the effects of bank’s CEO risk-taking incentives (Vega) on bank loan contracting In lending relationships, cumulative abnormal returns (CARs) in bank loan announcement studies is helpful in order to evaluate the firm performance(James 1987; Lummer and McConnell 1989; Dahiya et al., 2003; Billett et al., 1995; Billeett et al., 2006; Kang and Liu 2008) Various authors in their research show that positive announcement returns are observed in firms having low information asymmetry (Mikkelson and Partch 1986; James 1987; Lummer and McConnell 1989; Slovin et al., 1992; and Ross 2010) For example, Mikkelson and Partch (1986) and James (1987) argue that information embedded in the bank loan decisions reflect the health of firm to capital market by examining the positive excess returns associated with bank loan announcements Best et al., (1993) indicated that a positive CARs around the time of bank loan announcements can be considered as the signaling for banks’ valuable monitoring function Consistent with this idea, in this paper we evaluate the bank’s over-lending effect caused by CEO risk-taking incentive (Vega) is a good or bad signal by paying attention to the market response to bank loan announcement We attempt to answer the following five questions regarding the Vega effects on bank loan contracts: (i) Do banks with higher Vega earn lower cumulative abnormal returns around bank loan announcement date?; (ii) Do banks with higher Vega charge lower interest rates on loans?; (iii) Do Vega effects on bank loan contracts also exist in non-price terms(general covenants, financial covernants, collateral)?; (iv) Are Vega effects weaker by strong corporate governance mechanisms?; and (v) Do Vega effects still hold after adjusting for other CEO compensation schemes and CEO characteristics? We evaluate these questions by using a sample of 20,502 loans to 5,102 U.S firms between 1992 and 2014 We obtain all accounting variables and stock prices from the Compustat database Table CEO risk-taking incentives effect: Difference-in-Difference analysis This table reports the robustness checks on the effect of bank’s CEO risk-taking incentives on the loan spread using the difference-in-differences specification in the main result In panel A, we examine the treat×post effect of risk-taking incentives on the loan spread and CARs controlling two and four fixed effects Panel B shows parallel trend assumption for three years before and after the policy shocks We denote the banks that got bailed out as treatment unit (Treat 𝑖,𝑡−1 ), the rest of firms are considered as control units Also, we construct a dummy variable (Post 𝑖,𝑡 ) that equals one if a treatment firm’s fiscal year, t, falls after the year of Treat 𝑖,𝑡−1 , and zero otherwise The interaction term of variable (Treat 𝑖,𝑡−1 ) and ( Post 𝑖,𝑡 ) is denoted by “Treat×Post” variable B_1, B_2 and B_3 stand for one, two and three-year before the event; A_1, A_2 and A_3 stand for one, two and three-year after the event; C stand for current year of event The regression equation is as follows: Spread𝑖,𝑡 = 𝛼1 + 𝛼2 Treat × Post 𝑖,𝑡−1 𝑖,𝑡 + 𝛽 ′ 𝐹𝑖,𝑡−1 + 𝜃 ′ 𝑍𝑖,𝑡 + 𝜈𝑖 + 𝜇𝑡 + 𝜀𝑖,𝑡 𝐶𝐴𝑅[−5,5]𝑖,𝑡 = 𝛼1 + 𝛼2 Treat × Post 𝑖,𝑡−1 𝑖,𝑡 + 𝛽 ′ 𝐹𝑖,𝑡−1 + 𝜃 ′ 𝑍𝑖,𝑡 + 𝜈𝑖 + 𝜇𝑡 + 𝜀𝑖,𝑡 where 𝐶𝐴𝑅[−5,5]𝑖,𝑡 is the cumulative abnormal return (from the Fama-French four factor model) in the window [-5;5] the cumulative abnormal return of firm i in year t Spread𝑖,𝑡 represents the log of bank loan spread for firm i in year t; 𝐶𝐴𝑅𝑖,𝑡 is the cumulative abnormal return of firm i in year t ; 𝐹𝑖,𝑡−1 is a vector of control variables for borrowers and lenders i in year t-1;𝑍𝑖,𝑡 is a vector of control variables for loans i in year t.; 𝜈𝑖 and 𝜇𝑡 capture the CEO and year fixed effects, respectively; and 𝜀𝑖,𝑡 is the random error All continuous variables are winsorized at the 1st and 99th percentiles In all models, the t-statistics reported are based on heteroscedasticity and sample clustering at firm-level robust standard errors (White, 1980 and Petersen, 2009) *, **, and *** indicate statistical significance at the 10%, 5%, and 1% levels, respectively Variable definitions are provided in Appendix A Spread CAR[-5,5] Panel A (1) (2) (3) (4) Constant 2.2376** 1.1347 -0.0478 -0.0174 (2.12) (1.14) (-0.34) (-0.12) Treat×Post 0.1415** 0.1188** 0.0388*** 0.0353*** (2.47) (2.21) (3.91) (3.61) Control for CEO FE Year FE Loan purpose Loan type Controls Adj R2 Obs YES YES NO NO YES 0.5493 8680 YES YES YES YES YES 0.6197 8680 Spread Panel B Post B_3 B_2 B_1 C A_1 A_2 A_3 (1) 0.2458*** (25.63) 0.0664 (0.79) 0.0703 (0.51) -0.2078 (-0.84) -0.2701** (-2.26) 0.0872*** (4.56) 0.0255 (0.98) 0.0284 (0.56) (2) 1.1122*** (26.18) -0.0158 (-0.30) 0.1044 (1.47) -0.1430 (-1.37) -0.0794 (-0.67) 0.1433* (1.68) 0.1625*** (2.65) 0.1671*** (3.32) 43 YES YES NO NO YES 0.1738 3911 YES YES YES YES YES 0.1942 3911 CAR[-5,5] (3) (4) 0.0019* -0.0009 (1.73) (-0.18) -0.0079* -0.0038 (-1.91) (-0.86) -0.0027 -0.0009 (-0.87) (-0.33) -0.0012 0.0018 (-0.16) (0.24) -0.0211*** -0.0177*** (-4.13) (-3.18) -0.0460*** -0.0461*** (-4.38) (-4.20) -0.0021 0.0004 (-0.50) (0.12) -0.0063 -0.0038 (-1.18) (-0.64) Control for CEO FE Year FE Loan purpose Loan type Controls Adj R2 Obs NO YES NO NO NO 0.0844 8286 NO YES NO NO YES 0.4889 8286 44 NO YES NO NO NO 0.0426 4274 NO YES NO NO YES 0.0439 4283 Table CEO risk-taking incentives and bank loan spread: Evidence from bank governance channels This table presents the results of the effect of CEO risk-taking incentive on bank loan spread conditional on different corporate governance mechanisms The analysis follows the equation form: 𝑆𝑝𝑟𝑒𝑎𝑑𝑖,𝑡 = 𝛼0 + 𝛼1 𝑉𝑒𝑔𝑎𝑖,𝑡−1 + 𝛼2 𝑉𝑒𝑔𝑎𝑖,𝑡−1 × 𝐶𝐺𝑖,𝑡−1 + 𝛽 ′ 𝐹𝑖,𝑡−1 + 𝜃 ′ 𝑍𝑖,𝑡 + 𝛾𝑖 + 𝜇𝑡 + 𝜀𝑖,𝑡 The dependent variable Spreadi,t is the natural logarithm of bank loan spread for loan i in year t Vega𝑖,𝑡−1 is the change in the dollar value of CEO wealth for a 1% change in stock return volatility that a bank grants to its CEO in $million for bank i in year t-1 𝐶𝐺𝑖,𝑡−1 here are good corporate governance measures as high ratio of independent directors (ID_H), high institutional ownership (IO_H), middle board size (BS_M), high female (Fem_H), and high academic (Aca_H) 𝐹𝑖,𝑡−1 is a vector of control variables for borrowers and lenders i in year t − 1; 𝑍𝑖,𝑡 is a vector of control variables for loans i in year t; γi and μt represent the fixed effect of CEO and year respectively; and 𝜀𝑖,𝑡 is the random error In all models, the t-statistics reported are based on heteroscedasticity and sample clustering at firm-level robust standard errors (White, 1980 and Petersen, 2009) Superscripts *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively Constant Vega ID_H Vega × ID_H (1) 1.7099*** (4.21) -0.0045** (-2.19) -0.0329** (-2.38) 0.0045** (2.26) IO_H (2) 4.4096*** (3.88) -0.0289* (-1.94) Spread (3) 1.7266*** (3.14) -0.0056* (-1.86) (4) 1.6822*** (3.46) -0.0055*** (-3.38) -0.0660* (-1.92) 0.0131** (2.01) Vega × IO_H BS_M -0.0114* (-1.75) 0.0021** (2.17) Vega × BS_M Fem_H -0.0108 (-1.63) 0.0017** (2.10) Vega × Fem_H Aca_H -0.0285* (-1.80) 0.0063*** (3.35) Vega × Aca_H Control for CEO FE Year FE Loan purpose Loan type Controls Adj R2 Obs (5) 1.5194*** (3.76) 0.0000 (0.03) YES YES YES YES YES 0.5816 4358 YES YES YES YES YES 0.5869 5236 45 YES YES YES YES YES 0.5835 8298 YES YES YES YES YES 0.5663 5246 YES YES YES YES YES 0.4765 4333 Table Robustness check (I): Control for CEO characteristics and other CEO compensation This table presents ordinary least squares (OLS) regression results for the influences of CEO vega incentives on bank loan spread and cumulative abnormal return (CAR) by considering CEO characteristics and other CEO compensation as control variables The empirical model is: 𝑆𝑝𝑟𝑒𝑎𝑑𝑖,𝑡 = 𝛼0 + 𝛼1 𝑉𝑒𝑔𝑎𝑖,𝑡−1 + 𝛽 ′ 𝐹𝑖,𝑡−1 + 𝜃 ′ 𝑍𝑖,𝑡 + 𝛾𝑖 + 𝜇𝑡 + 𝜀𝑖𝑡 𝐶𝐴𝑅[−5,5]𝑖,𝑡 = 𝛼0 + 𝛼1 𝑉𝑒𝑔𝑎𝑖,𝑡−1 + 𝛽 ′ 𝐹𝑖,𝑡−1 + 𝜃 ′ 𝑍𝑖,𝑡 + 𝛾𝑖 + 𝜇𝑡 + 𝜀𝑖𝑡 where 𝑆𝑝𝑟𝑒𝑎𝑑𝑖,𝑡 represents the natural logarithm of bank loan spread for loan i in year t; 𝑉𝑒𝑔𝑎𝑖,𝑡−1 is the change in the dollar value of CEO wealth for 1% change in stockreturn volatility that a bank grants to its CEO in $million; 𝐶𝐴𝑅[−5,5]𝑖,𝑡 is the cumulative abnormal return (from the Fama-French four factor model) in the window [-5;5] the cumulative abnormal return of firm i in year t 𝐹𝑖,𝑡−1 is a vector of control variables for borrowers and lenders i in year t − 1; 𝑍𝑖,𝑡 is a vector of control variables for loans i in year t; γi and μt represent the fixed effect of CEO and year respectively; and 𝜀𝑖,𝑡 is the random error In all models, the t-statistics reported are based on heteroscedasticity and sample clustering at firm-level robust standard errors (White, 1980 and Petersen, 2009 Superscripts *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively To save space, we not report the coefficients for industry and year dummies Constant Vega Inside_debt (1) 1.9726 (0.93) -0.0248** (-2.31) 0.0321 (0.51) Tdc1 (2) 4.6388*** (4.40) -0.0252*** (-2.74) Spread (3) 4.6702*** (4.32) -0.0280*** (-2.93) (5) 0.1636 (0.61) -0.0063*** (-2.93) 0.0018 (0.22) -0.0171 (-1.37) Delta CAR[-5,5] (6) (7) -0.1031 -0.1049 (-0.72) (-0.73) -0.0054*** -0.0056*** (-3.09) (-3.27) 0.0000 (0.56) 0.0000 (0.61) -0.2512* (-1.91) YES YES YES YES YES YES YES YES YES YES (8) -0.0495 (-0.16) -0.0039* (-1.74) 0.0020 (0.83) Tenure Control for CEO FE Year FE Loan purpose Loan type Controls (4) 6.7157*** (6.99) -0.0244** (-2.16) YES YES YES YES YES YES YES YES YES YES 46 0.0031 (0.14) YES YES YES YES YES YES YES YES YES YES YES YES YES YES YES YES YES YES YES YES Adj R2 Obs 0.5332 3139 0.5940 8298 0.5956 8182 0.6032 5715 47 0.0605 2227 0.0495 4283 0.0437 4199 0.0734 2244 Table 10 Robustness check (II): Bank size analysis and the change regression This table presents the results of the effect of bank CEO vega incentives on different bank sizes and presents the change regression results of the effect of bank CEO risk-taking incentives (Vega) Model (1) follows the below equation form: 𝑆𝑝𝑟𝑒𝑎𝑑𝑖,𝑡 = 𝛼0 + 𝛼1 𝑉𝑒𝑔𝑎𝑖,𝑡−1 + 𝛼2 𝑉𝑒𝑔𝑎𝑖,𝑡−1 × 𝑆𝑚𝑎𝑙𝑙 𝑏𝑎𝑛𝑘𝑖,𝑡−1 + 𝛼3 𝑆𝑚𝑎𝑙𝑙 𝑏𝑎𝑛𝑘𝑖,𝑡−1 + 𝛽 ′ 𝐹𝑖,𝑡−1 + 𝜃 ′ 𝑍𝑖,𝑡 + 𝛾𝑖 + 𝜇𝑡 + 𝜀𝑖,𝑡 where 𝑆𝑝𝑟𝑒𝑎𝑑𝑖,𝑡 represents the natural logarithm of bank loan spread for loan i in year t; 𝑉𝑒𝑔𝑎𝑖,𝑡−1 is the change in the dollar value of CEO wealth for a 1% change in stock-return volatility that a bank grants to its CEO in $million; Small bank 𝑖,𝑡−1 is dummy variable that equals if the bank size is smaller than the sample median and otherwise; 𝐹𝑖,𝑡−1 is a vector of control variables for borrowers and lenders i in year t − 1; 𝑍𝑖,𝑡 is a vector of control variables for loans and macroeconomic factors i in year t γi and μt represent the fixed effect of industry and year respectively; and 𝜀𝑖,𝑡 is the random error Model (2) follows the below equation form: ∆𝑆𝑝𝑟𝑒𝑎𝑑𝑖,𝑡 = 𝛼0 + 𝛼1 ∆𝑉𝑒𝑔𝑎𝑖,𝑡−1 + 𝛽 ′ ∆𝐹𝑖𝑟𝑚𝑖,𝑡−1 + 𝜃 ′ 𝑍𝑖,𝑡 + 𝜀𝑖,𝑡 where the dependent variable ∆𝑆𝑝𝑟𝑒𝑎𝑑𝑖,𝑡 is the change in loan spread for firm i in year t from year t-1 ∆𝑉𝑒𝑔𝑎𝑖,𝑡−1 represents the change in bank CEO risk-taking incentives, Vega i in year t-1 from year t-2 ∆𝐹𝑖𝑟𝑚𝑖,𝑡−1 represents the change in firm characteristics for lenders and borrowers i in year t-1 from year t-2 𝑍𝑖,𝑡 is the vector of the control variables for loan and macroeconomic factors i in year t In all specifications, the t-statistics reported are based on heteroscedasticity (White, 1980) The sample period is 1992-2014 All the variables are defined in Appendix A *, ** and *** denote significance levels of 10%, 5% and 1%, respectively Constant Vega Vega×Small bank Small bank (1) 5.1862*** (9.99) -0.0099 (-1.38) -0.0337** (-2.39) 0.2581** (2.52) ΔVega Assets Leverage Tangibility Profitability MB Z score CF volatility CEO age Maturity Loansize Performance GenCov (2) 1.9273** (2.32) -0.0525** (-2.18) -0.1151*** (-16.47) 0.8313*** (24.26) -0.2197*** (-5.35) -0.9530*** (-5.89) -0.2101*** (-5.15) 0.0100*** (2.96) 0.0001 (0.86) 0.2494*** (2.78) -0.0379** (-2.35) -0.0770*** (-10.23) -0.0798*** (-5.48) 0.0888*** 48 0.0031 (0.02) -0.0323 (-0.72) 0.0033 (0.27) -0.0331 (-0.83) -0.0064 FinCov Credit spread Term spread L_assets L_leverage L_loandep (7.20) 0.0413** (2.44) -0.2294* (-1.69) 0.0753*** (3.47) 0.0213* (1.77) 0.0016 (0.39) -0.0839** (-2.08) ΔAssets -0.0223 (-0.36) 0.6743*** (3.95) 0.0787 (0.33) -0.3450 (-1.44) -0.0776 (-0.67) 0.0010 (0.06) -0.0006*** (-3.67) -0.0218 (-0.12) -0.0288** (-2.43) -0.1611 (-0.67) ΔLeverage ΔTangibility ΔProfitability ΔMB ΔZ-score ΔCF-volatility Δ L_assets ΔL_leverage ΔL_loandep Control for Industry FE Year FE Loan purpose Loan type Obs Adj R2 (-0.19) 0.0322 (0.67) 0.6825** (2.32) -0.0243 (-0.59) Yes Yes Yes Yes 8,362 0.6458 49 Yes Yes Yes Yes 1,556 0.1966 Table 11 Robustness check(III): Bank’s CEO risk-taking incentives and cumulative abnormal returns These tables present the ordinary least squares (OLS) regression results of bank CEO risk-taking incentives on bank loan spread 𝐶𝐴𝑅[−3,3]𝑖,𝑡 = 𝛼0 + 𝛼1 𝑉𝑒𝑔𝑎𝑖,𝑡−1 + 𝛽 ′ 𝐹𝑖,𝑡−1 + 𝜃 ′ 𝑍𝑖,𝑡 + 𝛾𝑖 + 𝜇𝑡 + 𝜀𝑖𝑡 Where 𝐶𝐴𝑅[−3,3]𝑖,𝑡 is the cumulative abnormal return (from the Fama-French four factor model) in the window [-3;3] the cumulative abnormal return of firm i in year t 𝑉𝑒𝑔𝑎𝑖,𝑡−1 represents the incentives for the CEO of bank i in year t-1; 𝐹𝑖,𝑡−1 is a vector of control variables for borrowers and lenders i in year t-1, including borrower and lender characteristics; 𝑍𝑖,𝑡 is a vector of control variables for loans i in year t γi and μt represent the fixed effect of CEO and year, respectively In all models, the t-statistics reported are based on heteroscedasticity and sample clustering at firm-level robust standard errors (White, 1980 and Petersen, 2009) The sample period is 1992-2014 All the variables are defined in Appendix A *, ** and *** denote the significance level of 10%, 5% and 1%, respectively Constant Vega L_asset L_leverage L_loandep CEO age Maturity Loansize Performance FinCov GenCov Leverage Tangibility Profitability MB Z_score CF_volatility CAR[-3,3] (2) 0.0012 (0.01) -0.0050*** (-3.77) 0.0008 (0.12) -0.0000 (-0.00) -0.0120 (-0.98) 0.0013 (0.77) -0.0007 (-0.48) 0.0001 (0.25) 0.0001 (0.08) 0.0007 (0.38) -0.0007 (-0.52) -0.0029 (-0.94) -0.0021 (-0.92) 0.0103* (1.73) -0.0005 (-0.15) -0.0003* (-1.92) 0.0000 (0.51) (1) -0.0063 (-0.06) -0.0049*** (-3.75) -0.0006 (-0.09) 0.0000 (0.05) -0.0097 (-0.80) 0.0017 (1.00) -0.0006 (-0.61) 0.0003 (0.66) 0.0001 (0.06) 0.0005 (0.28) -0.0007 (-0.54) -0.0023 (-0.80) -0.0025 (-1.08) 0.0108* (1.83) -0.0001 (-0.04) -0.0003* (-1.93) 0.0000 (0.62) Control for 50 (3) -0.0094 (-0.09) -0.0049*** (-3.73) 0.0012 (0.20) 0.0000 (0.05) -0.0131 (-1.07) 0.0014 (0.81) -0.0000 (-0.03) 0.0002 (0.39) -0.0002 (-0.10) 0.0005 (0.28) -0.0008 (-0.60) -0.0024 (-0.80) -0.0027 (-1.15) 0.0099* (1.65) 0.0015 (0.45) -0.0003* (-1.88) 0.0000 (0.57) CEO FE Year FE Loan purpose Loan type Adj R2 Obs YES YES NO NO 0.0359 4274 YES YES YES NO 0.0347 4274 51 YES YES YES YES 0.0335 4274 Appendix A: Variable definition Variables Definition Data source Panel A: CEO risk-taking incentive Vega Nature log of the dollar amount change of CEO stock and option portfolio per percent change in standard deviation of the annualized stock return (US$ millions) Execucomp Panel B: Loan announcement returns CAR[-5,5] Sum of abnormal returns on the event windows from days before the event to following days CRSP & Dealsan CAR[-3,3] Sum of abnormal returns on the event windows from days before the event to following days CRSP &Dealsan Panel C: Loan characteristics Spread Natural logarithm of loan spread DealScan Maturity Natural logarithm of loan maturity in months DealScan Loan size Natural logarithm of loan amount US$ millions DealScan Performance Dummy variable, equal to one if a loan facility uses performance pricing, and zero otherwise DealScan Collateral Dummy variable, equal to one if a loan is secured, and zero otherwise DealScan GenCov Number of general covenants DealScan FinCov Number of financial covenants DealScan Loan type Dummy variable for loan types, including term loan, revolver greater than one year, revolver less than year, and 364-day facility DealScan Loan purpose Dummy variable for loan purposes, including corporate purposes, debt repayment, working capital, takeover, etc DealScan Panel D: Firm characteristics Leverage Long-term debt and debt in current liabilities divided by total assets Compustat Tangibility Net property, plant, and equipment divided by total assets Compustat Profitability Earnings before interest, taxes, depreciation, and amortization (EBITDA), divided by total assets Compustat MB Market value of net assets to book value of net assets ratio Compustat and CRSP Z-score Modified Altman’s Z-score (1.2 × working capital + 1.4 × retained earnings + 3.3×EBIT + 0.999×sales)/Total assets Compustat CF-volatility Standard deviation of quarterly cash-flows from operations over the four fiscal years prior to the loan initiation year, scaled by total assets Compustat Panel E: Lender characteristics L_Asset L_ Leverage Log of total assets Compustat Bank (CB) Ratio of assets to book value of equity Compustat Bank (CB) 52 L_loandep Ratio of average balance of loans to average balance of deposits Compustat Bank (CB) Panel F: CEO compensation and characteristics CEO age Natural logarithm of Age of CEO Age of the CEO when the company signs the bank loan contract Execucomp CEO tenure Natural logarithm of CEO tenure Number of years the CEO held his/her position in a company before he/she signed the bank loan contract Execucomp TDC1 We follow the definition of TDC1 in ExecuComp to calculate this variable TDC1 includes salary, bonus, stock awards, option awards, long-term incentive plans, and other annual compensation such as perquisites and other personal benefits (including termination or change-in-control payments, 401K plans, etc) (US$ millions) Execucomp CEO inside debt ratio Natural logarithm of CEO D/E (the value of inside debt divided by the total value of shares and options owned We define the inside debt of the CEO as the sum of the balance in the CEO’s pension fund and nonqualified deferred compensation Execucomp and SEC Edgar DEF 14A Filings Delta Natural logarithm of dollar change in wealth associated with a 1% change in the firm’s stock price (US$ million) Execucomp Panel G: Bank governance mechanisms Independent Directors (ID) Percentage of outside directors Risk Metrics Institutional Ownership (IO) Percentage of shares held by institutional investors in the bank Thomson Reuters Board size(BS) Number of board directors Risk Metrics Female(Fem) The proportion of female directors in the board Risk Metrics Academic(Aca) A dummy variable that equals one if the board includes at least one director from academia, and zero otherwise * BoardEx CRSP: Center for Research in Security Prices; BSS (2016) is Berg, Saunder and Steffen (2016); DLM (2012) is Demerjian, Lev and McVay (2012) 53 Appendix B: Sample banks 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 AMSOUTH BANCORPORATION BANK ONE CORP BANK OF HAWAII CORP BANKBOSTON CORP BANK OF NEW YORK MELLON CORP BANKERS TRUST CORP BARNETT BANKS INC BBVA COMPASS BANCSHARES INC CHASE MANHATTAN CORP -OLD JPMORGAN CHASE & CO CITICORP CITY NATIONAL CORP COMERICA INC COMMERCE BANCSHARES INC CONTINENTAL BANK CORP CORESTATES FINANCIAL CORP CULLEN/FROST BANKERS INC DAUPHIN DEPOSIT CORP FIFTH THIRD BANCORP REGIONS FINANCIAL CORP FIRST AMERICAN CORP/TN FIRST OF AMERICA BANK CORP FIRST CHICAGO CORP M & T BANK CORP FIRST FIDELITY BANCORP FIRST INTERSTATE BNCP U S BANCORP WACHOVIA CORP HIBERNIA CORP -CL A HUNTINGTON BANCSHARES 54 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 MELLON FINANCIAL CORP MERCANTILE BANCORPORATION BANK OF AMERICA CORP NATIONAL CITY CORP NORTHERN TRUST CORP WELLS FARGO & CO PNC FINANCIAL SVCS GROUP INC RIGGS NATIONAL CORP KEYCORP SOUTHTRUST CORP STATE STREET CORP SUNTRUST BANKS INC CRESTAR FINANCIAL CORP WILMINGTON TRUST CORP ASSOCIATED BANC-CORP FIRST MIDWEST BANCORP INC SYNOVUS FINANCIAL CORP MARK TWAIN BANCSHARES FIRST COMMERCIAL CORP COLONIAL BANCGROUP WASHINGTON MUTUAL INC PEOPLE'S UNITED FINL INC TD BANKNORTH INC PROVIDENT BANKSHARES CORP SVB FINANCIAL GROUP SOUTH FINANCIAL GROUP INC FIRST COMMONWLTH FINL CP/PA FIRST NIAGARA FINANCIAL GRP PROSPERITY BANCSHARES INC Appendix C: Vega measure Following Core and Guay (2002), Coles, Daniel and Naveen (2006), we first define vega as the dollar amount change in the CEO stock and option portfolio per onepercent change in the standard deviation of the annualized stock return (stock-return volatility) Then, to generate annual estimates of Vega, for fiscal years 1992-2005 we use the “one-year approximation” method, which explains about 99 percent of the variation in option portfolio values and sensitivities that one would obtain from having full information about option portfolios For fiscal year 2007 and later, however, it is unnecessary to use one year approximation (OA) in Core and Guay (2002), because all firms on Execucomp disclose information with complete portfolios of exercised and unexercised stock options in the outstanding equity tables using the new-format rules for DEF14A filings.9 For 2006, 84 percent of firms report using the new format, while the rest report under the old format For calculation of vega under the new reporting format, Execucomp provides option data in the current year and previously granted options, which include the number of options granted, option exercise price, time to maturity, the firm’s expected dividend yield, the firm’s expected stock return volatility, and the risk-free interest rate for that year 10 All the above variables are inputs to estimate option values and sensitivities from the formulas in the Black–Scholes option-valuation model (1973), modified by Merton (1973) to account for dividends: Option value =[S𝑒 −𝑑𝑡 𝑁(𝑍) − 𝑋𝑒 −𝑟𝑡 𝑁(𝑍 − 𝜎𝑇 (1/2) )] where Z is [ln(S/X)+T(r-d+𝜎 /2)]𝜎𝑇 (1/2) ; N is the cumulative probability function for normal distribution; S is stock price; X is the option-exercise price; 𝜎 is the expected stock-return volatility; r is the natural logarithm of risk-free interest rate; T is option time to maturity, in years; d is the natural logarithm of dividend yield The sensitivity of option value with respect to a 0.01 change in stock-return These data are available from the “Outstanding Equity Awards” table in the WRDS interface for Execucomp and in “outstandingawards.sas7bdat” if the data are downloaded from the WRDS server (/wrds/comp/sasdata/execcomp) We calculate this because Execucomp stopped providing this variable (BS_VOLATILITY) as of 2006 We follow the Execucomp methodology as closely as possible Accordingly, we (i) use annualized standard deviation of stock returns estimated over the 60 months prior to the beginning of the fiscal period; (ii) require at least 12 months of returns data; (iii) use mean volatility (across all firms) for that year if 12 months of data are not available; and (iv) winsorize the volatility estimates at the 5𝑡ℎ and 95𝑡ℎ levels 55 volatility (where 𝑁 ′ is the normal density function) is: [𝜕(𝑜𝑝𝑡𝑖𝑜𝑛 𝑣𝑎𝑙𝑢𝑒)/𝜕(𝑠𝑡𝑜𝑐𝑘 𝑟𝑒𝑡𝑢𝑟𝑛 𝑣𝑜𝑙𝑎𝑡𝑖𝑙𝑖𝑡𝑦)] ∗ 0.01 = 𝑒 −𝑑𝑡 𝑁 ′ (𝑍)𝑆𝑇 (2) × 0.01 The vega of all vested and unvested options awards are summed for each executive-year to obtain our vega of option portfolio To calculate vega with pre-2006 data, since Execucomp only provides firms’ detailed reports for the current year’s option grants, we therefore use the one-year approximation method in Core and Guay (2002) to estimate data on previously granted options We consider three option portfolios: the current year’s option grants, the portfolio of unvested options from previously granted awards, and the portfolio of vested options CEOs’ incentives are given by the summation of the incentives from these three portfolios For the current year’s option grants, we obtain the number of options granted during that year, the stated exercise price, and maturity (based on expiration date), along with the firm’s expected dividend yield, the firm’s expected stock-return volatility, and the risk-free interest rate for that year, all from Execucomp, are used to estimate option values and sensitivities from formulas described earlier To calculate the portfolio of previously granted unvested options, we make assumptions based on exercise price and times to maturity as in Core and Guay (2002) First, we estimate the total number of options in the portfolio and the average exercise price of each option We net off the total number of options granted in the current year from the number of unvested options to estimate the number of previously granted unvested options We then calculate the difference between the reported intrinsic value of all unvested options and the intrinsic value of the current year’s grants Finally, we subtract the average intrinsic value of each option in the portfolio from the stock price to obtain the average exercise price of each previously granted unvested option If a firm grants options in the current year, the time to maturity of previously granted unvested options is set to the actual maturity of current year option grants minus one If no option grants are made in the current year, we assume that the average maturity of previously granted unvested options is nine years For vested options, we calculate the average exercise price based on the realizable value and number of vested options The maturity of vested options equals the maturity 56 of unvested options minus three The vega is the sum of the vega of the current year options plus previously granted options (both vested and unvested) 57 ...銀行經理人風險承擔動機的黑暗面：從銀行放款決策分析之 The Dark Side of Bank CEO Risk- taking Incentives: Evidence from Bank Lending Decisions 研 究 生: 陳氏垂玲 Student: TRAN THI THUY LINH 指 導 教 授: 駱建陵 Advisors: Prof CHIEN-LING LO 林智勇 Prof CHIH-YUNG... iii The Dark Side of Bank CEO Risk- taking Incentives: Evidence from Bank Lending Decisions Student: Tran Thi Thuy Linh Advisors: Chien-Ling Lo Chih-Yung Lin Doctor of Philosophy Program Major of. .. 17 4.1 Bank? ??s CEO risk- taking incentives and cumulative abnormal returns(CARs) 17 4.2 Bank? ??s CEO risk- taking incentives and bank loan spread 17 4.3 Bank? ??s CEO risk- taking incentives