CEO Overconfidence & Financial Crisis: Evidence From Bank Lending & Leverage
Po-Hsin Ho
Department of Business Administration
College of Business
National Taipei University
New Taipei City 23741, Taiwan
Chia-Wei Huang
College of Management
Yuan Ze University
Taoyuan 32003, Taiwan
Chih-Yung Lin
College of Management &
Innovation Center for Big Data and Digital Convergence
Yuan Ze University
Taoyuan 32003, Taiwan
Ju-Fang Yen
Department of Statistics
College of Business
National Taipei University
New Taipei City 23741, Taiwan
September 5, 2015
Forthcoming in Journal of Financial Economics
Abstract
Over a period that includes the 1998 Russian crisis and 2007–2009 financial crisis, banks with overconfident chief executive officers (CEOs) were more likely to weaken lending standards and increase leverage than other banks in advance of a crisis, making them more vulnerable to the shock of the crisis. During crisis years, they generally experienced more increases in loan defaults, greater drops in operating and stock return performance, greater increases in expected default probability, and higher likelihood of CEO turnover or failure than other banks. CEO overconfidence thus can explain the cross-sectional heterogeneity in risk-taking behavior among banks.
CEO Overconfidence & Financial Crisis: Evidence From Bank Lending & Leverage – Introduction
The near-decade before the US credit market freeze in the fall of 2007 was a period of unprecedented prosperity and credit expansion. In a speech delivered on March 7, 2001, Federal Reserve System chairman Alan Greenspan pointed out that “there is doubtless an unfortunate tendency among some, I hesitate to say most, bankers to lend aggressively at the peak of a cycle and that is when the vast majority of bad loans are made.”
Greenspan cautioned against an overly optimistic assessment of borrower prospects during a credit boom. Unrealistic assumptions make banks more vulnerable when a credit boom is followed by a crisis. We examine whether banks with an overconfident attitude acted differently from other banks in terms of lending standards and bank leverage prior to a crisis and then how such banks performed during crisis years. We focus on the top decision maker in the bank, the chief executive officer (CEO).
Because CEOs are the primary influence on bank financing and investment decisions, CEO attitudes toward borrower prospects could affect their banks’ lending standards and leverage levels, which then can affect a bank’s sensitivity to a crisis. We specify that banks with an overconfident attitude are banks with overconfident CEOs, who generally think they are better than they are in terms of skill and judgment or in gauging the prospects of a successful outcome. We thus examine whether CEO overconfidence can help explain a bank’s risk taking before a crisis and the poor performance of a bank during a crisis.
Managerial overconfidence can influence bank risk-taking behaviors in several ways. Hirshleifer and Luo (2001), Malmendier and Tate (2008), and Gervais, Heaton, and Odean (2011) show that overconfident CEOs overestimate the probability of a positive state and the likelihood of returns to be generated from an investment project. They underestimate the downside risk of a project and tend to choose a project with higher true risks than optimal.3 During an economic upswing, an overconfident CEO who is more bullish than others on prospects for the economy could relax lending standards and increase bank leverage more than other banks while they take exposures believed to be the most profitable for current shareholders. Yet, by taking greater risk, overconfident CEOs make their banks more vulnerable to an external shock such as a financial crisis.
We collect CEO overconfidence data from publicly listed US banks over 1994–2009, a period that includes the 1998 Russian financial crisis and the most recent worldwide financial crisis of 2007–2009. Both crises followed a period of lending growth (Ivashina and Scharfstein, 2010; Becker and Ivashina, 2014), and both are among the worst financial crises in the last 50 years (Fahlenbrach, Prilmeier, and Stulz, 2012). Fahlenbrach, Prilmeier, and Stulz (2012) find that some bank characteristics connoting risk-taking behaviors are significantly associated with poor performance in both crises. Hence, if overconfidence as a managerial trait can explain heterogeneity in risk-taking behavior among banks, an overconfident bank should have suffered more than other banks in the two crises. Following Fahlenbrach, Prilmeier, and Stulz (2012), we take 1998 as the start of the Russian crisis and 2007 as the start of the most recent financial crisis. We then define 1998 and 2007–2009 as crisis years and other years as non-crisis years.
We use a stock options-based proxy for CEO overconfidence and construct our measure using Standard & Poor’s ExecuComp database. Following Campbell, Gallmeyer, Johnson, Rutherford, and Stanley (2011), we classify CEOs who postpone exercising stock options that are more than 100% in the money at least twice during their tenure as overconfident (from the first time a CEO is seen to postpone exercise). The rationale is that a manager who chooses to keep holding deep-in-the-money stock options after the vesting period is likely to be overconfident about the firm’s future prospects.
See full PDF below.
