Insider versus outsider CEOs, executive compensation, and accounting manipulation

https://doi.org/10.1016/j.jacceco.2017.01.002Get rights and content

Abstract

This paper examines the role of the financial reporting environment in selecting a new CEO from within versus outside the organization. Weak reporting controls allow the CEO to misreport performance information, which reduces the board's ability to detect and replace poorly-performing CEOs as well as aggravates incentive contracting. We show that these adverse effects are stronger when the CEO is an outsider rather than an insider. Our model predicts that boards are more likely to recruit a CEO from the outside when the performance measures with which the new hire is assessed are harder to manipulate.

Introduction

When searching for a new CEO, corporate boards face an important question: Should they recruit a CEO from outside the organization or promote someone from within? Outsiders are typically considered to be more risky than insiders because corporate boards have less information about outsiders' strengths, experiences, and leadership style than they have about internal candidates. In addition, outsiders are less familiar with the organization's unique culture and inner workings. Outsider CEOs can nevertheless be valuable to the firm because they bring new ideas and fresh perspectives and are generally more open to transformational changes than insider CEOs.1 One implication of this argument is that boards tend to promote internal candidates when the continuation of the current strategy and culture is desirable, but prefer external candidates when major changes are required (e.g., Zajac, 1990, Parrino, 1997, Farrell and Whidbee, 2003).2

The notion that bringing in an outsider is more risky than promoting an insider gives rise to another factor relevant for the selection decision – the firm's financial reporting environment. Financial reporting plays an important role because boards use earnings information not only for incentive contracting, but also for assessing how well the new CEO matches the needs of the organization and deciding whether to retain or replace him. We show that incentive problems and the board's ability to assess and replace poorly-performing CEOs influence the board's initial decision over what type of CEO to hire. Our model generates new empirical predictions regarding the determinants of CEO selection and the effects of this decision on: optimal contracting, expected CEO compensation, the extent of earnings manipulation, and the frequency of forced CEO turnover.3

The heightened risks associated with bringing in an outsider do not necessarily put outside candidates at a disadvantage. As Lazear (1998) and Hermalin (2005) point out, boards have an option to replace the new hire if he turns out to be the wrong person for the job. Indeed, it is not uncommon that a CEO is fired within the first two years of his tenure.4 Boards, and the shareholders they represent, benefit from the outsiders’ upside potential and can limit downside risk by making appropriate subsequent replacement decisions. However, this argument relies critically on the board's ability to identify and dismiss CEOs who perform poorly.

Empirical evidence by Weisbach (1988) and Murphy and Zimmerman (1993) suggests that the firm's financial reporting system is an important source of information for assessing and replacing executives.5 If, however, CEOs can manipulate accounting reports, the board may be unable to identify and correct a poor hiring decision. Consequently, when CEOs can more easily manipulate performance information (for example, due to weaker reporting controls), the probability of CEO turnover declines and the board's option to replace the incumbent becomes less valuable. Importantly, we show that the effects of weaker reporting controls on CEO turnover and the replacement option value are greater for CEOs hired from the outside than for those promoted from within. There are two reasons behind this result and both are driven by the assumption that outsiders pose a greater downside risk. First, the reduced ability to identify and replace unsuccessful CEOs due to distorted performance information is a bigger problem when the CEO is from the outside, because outsiders are more likely to fail. Second, as we discuss further below, outsiders have stronger incentives to manipulate the accounting report than insiders, and the difference in manipulation incentives between the two types of candidates further increases as reporting controls become weaker. In fact, for sufficiently weak reporting controls, the outsider's manipulation incentive exceeds that of the insider by a wide margin such that the probability of early dismissal and the value of the replacement option are both smaller for the outsider than for the insider despite the outsider's greater downside risk.

The reporting system affects the desirability of outsiders versus insiders also through its impact on incentive contracting. To encourage the CEO to take productive but personally costly actions, the board awards the CEO a bonus for high reported performance.6 Because outsiders have a greater downside risk, they are less likely to succeed despite high effort. Outsiders must therefore receive a higher bonus to have sufficient incentives to work hard. The higher bonus, in turn, creates stronger incentives to manipulate the report, which increases the cost of the incentive system. As reporting controls become weaker, manipulation and the cost of the incentive contract increase for both types of candidates, but more quickly for outsiders than for insiders.

These arguments show that outsider CEOs have advantages and disadvantages relative to insider CEOs. On the one hand, outsiders have a higher option value as long as reporting controls are sufficiently strong. On the other hand, the incentive problem is more severe for outsiders, which implies higher expected compensation (and CEO rents). As reporting controls improve, option value increases and the expected compensation decreases for both types of candidates, but these effects are stronger for outsiders than for insiders. Thus, tighter reporting controls increase the advantages and decrease the disadvantages of outside candidates relative to inside candidates.

The analysis generates a number of new empirical predictions. First, boards are more likely to hire outsider CEOs in firms or countries with stronger reporting controls. Second, outsider CEOs engage in more earnings manipulation, face steeper incentive pay, and obtain higher expected compensation than insider CEOs, and these differences are greater when reporting controls are weaker. Third, outsiders have a shorter expected tenure relative to insiders if reporting controls are strong, and the reverse holds if reporting controls are weak.

Other models of CEO selection include Hermalin, 2005, Murphy and Zábojník, 2004 2007), and Palomino and Peyrache (2013). Murphy and Zábojník (2004, 2007) argue that changes in the economic environment raise the value of general managerial skills relative to firm-specific skills, which in turn increases the desirability of outsider CEOs. When competition in the managerial labor market is high, the increased demand for outsiders translates into higher executive compensation. Building on Murphy and Zábojník (2004, Murphy and Zábojník, 2007, Palomino and Peyrache, 2013 consider a setting in which outsiders have pre-contract private information about their firm-specific skills, whereas the skills of insiders are commonly known. The additional information asymmetry leads to greater expected compensation for CEOs hired externally, relative to those promoted internally. In contrast, in our setting, the difference in expected compensation between insiders and outsiders is driven by the effort incentive problem and the scope to manipulate performance measures. Hermalin (2005) studies the value of outsiders versus insiders in a setting in which the board engages in costly information acquisition to uncover the incumbent's ability before making the replacement decision. He demonstrates that outside candidates have an advantage over inside candidates due to their greater replacement option value. As discussed above, we show that the value of the replacement option declines when the CEO can distort the performance measure on which the replacement decision is based and this effect is stronger for outsiders than for insiders. Our model also differs from Hermalin (2005) in that we consider optimal incentive contracting. In sum, our paper contributes to the extant literature by studying the effects of the financial reporting environment on the desirability of appointing a new CEO from within versus outside the firm, and by examining how this decision affects the optimal incentive contract, expected compensation, the extent of manipulation, and the frequency of CEO replacement.

The paper also adds to the literature on CEO turnover, which includes the work by Cremer, 1995, Arya et al., 1998, Almazan and Suarez, 2003, Laux, 2008, and  Inderst and Mueller (2010). Among these studies, only Arya et al. (1998) consider how changes in the reporting system affect forced turnover.7 They show that allowing the manager to destroy earnings information through manipulation can be optimal because it enables the board to credibly commit not to fire the incumbent at an interim stage, which reduces the compensation required to induce the CEO to join the firm ex ante. In contrast, in our model, manipulation is always undesirable because it leads to both less efficient replacement decisions and higher CEO compensation. In addition, our study differs from the extant work on turnover because we address the costs and benefits of appointing insider versus outsider CEOs.

In Section 2, we outline the model. In 3 Optimal contract and manipulation incentives, 4 Economic output and option value, we analyze how the origin of the CEO (insider versus outsider) affects the optimal contract, the cost of compensation, the level of manipulation, the probability of CEO dismissal, the ex ante value of the option to replace the incumbent, and how the reporting environment interacts with these effects. In Section 5, we consider the board's choice of whether to appoint an outsider versus an insider. We discuss the empirical implications of the model in Section 6 and conclude in Section 7. All proofs are in Appendix A.

Section snippets

The model

We consider a game with two risk-neutral players: a CEO and a benevolent board of directors that acts in the best interest of shareholders.8

Timing: There are three dates. At date 0, the board hires a CEO to run the firm and offers him an incentive pay plan. After signing the contract, the CEO takes a personally costly action that increases expected output. At

Optimal contract and manipulation incentives

In this section, we determine the optimal contract and the CEO's equilibrium manipulation choice when the board hires a new CEO of origin n. For convenience, we ignore the superscript n when we refer to an arbitrary CEO and use the superscript I or O when we refer to a particular CEO.

We start the analysis with the CEO's optimal choice of manipulation. When the accounting signal is high, the CEO has no reason to manipulate the report because in the optimal contract wH>wL (see below). When the

Economic output and option value

We now consider how the accounting environment affects the firm's expected output. Expected output can be expressed asE+Y,where E is the output when the board cannot replace the incumbent (defined in (1)) andY=(1aHp)(1m*)(AxL),is the ex ante value of the replacement option. The option value Y is the product of the probability that the board replaces the incumbent, π(1aHp)(1m*), which only occurs when the report is unfavorable, and the associated marginal increase in firm value (AxL)>0.

Insider versus outsider

We are now ready to study the conditions under which the board prefers an outside candidate over an inside candidate. Expected firm value, denoted V, is the difference of the expected output given in (12) and expected CEO compensation given in (10):V=E+Y(g)C(g).

The board recruits a CEO from the outside if and only if VOVI, which can be rewritten in terms of the outsider's output potential:xHOx^HOEI+(YIYO)(CICO)(1aHpO)xLOaHpO.

Holding xHI,xLI, and xLO constant, there exists a threshold,

Determinants of CEO selection

A number of empirical studies have looked at factors that are associated with the selection of insider versus outsider CEOs. For example, evidence suggests that boards are more likely to recruit CEOs from the outside when changes in the direction of the firm are desirable (Parrino, 1997, Farrell and Whidbee, 2003), when the industry is more homogenous (Parrino, 1997, Zhang and Rajagopalan, 2003), when the proportion of outside directors sitting on the board is greater (Borokhovich et al., 1996

Conclusion

We analyze how the financial reporting environment affects the benefits and costs of appointing a new CEO from within versus outside the organization. Although our analysis is framed in terms of CEO selection, the results carry over to the more general question of whether to fill a senior-level job opening (e.g., department or store manager, vice president, or college dean) with an external or internal candidate. Going outside is typically viewed as more risky because employers know less about

References (36)

  • A. Almazan et al.

    Entrenchment and severance pay in optimal governance structures

    J. Financ.

    (2003)
  • C. Armstrong et al.

    The role of information and financial reporting in corporate governance and debt contracting

    J. Account. Econ.

    (2010)
  • A. Arya et al.

    Abandonment options and information system design

    Rev. Account. Stud.

    (2003)
  • A. Arya et al.

    Earnings management and the revelation principle

    Rev. Account. Stud.

    (1998)
  • Battley, S., 2012. Outsider or insider CEO: why Yahoo and Citigroup leaders have different success odds....
  • K. Borokhovich et al.

    Outside directors and CEO selection

    J. Financ. Qual. Anal.

    (1996)
  • J. Brickley et al.

    Corporate governance myths: comments on Armstrong, Guay, and Weber

    J. Account. Econ.

    (2010)
  • J. Cremer

    Arm's length relationships

    Q. J. Econ.

    (1995)
  • K. Crocker et al.

    The economics of earnings manipulation and managerial compensation

    RAND J. Econ.

    (2007)
  • D. Dalton et al.

    Inside/outside succession and organizational size: the pragmatics of executive replacement

    Acad. Manag. J.

    (1983)
  • S. Dutta et al.

    Earnings management and dynamic incentives

    (2015)
  • S. Dutta et al.

    The effect of earnings forecasts on earnings management

    J. Account. Res.

    (2002)
  • R. Dye

    Earnings management in an overlapping generations model

    J. Account. Res.

    (1988)
  • K.A. Farrell et al.

    Impact of firm performance expectations on CEO turnover and replacement decisions

    J. Account. Econ.

    (2003)
  • G.A. Feltham et al.

    Performance measure congruity and diversity in multi-task principal/agent relations

    Account. Rev.

    (1994)
  • E. Goldman et al.

    An equilibrium model of incentive contracts in the presence of information manipulation

    J. Financ. Econ.

    (2006)
  • J. Guthrie et al.

    Contextual influences on executive selection: firm characteristics and CEO experience

    J. Manag. Stud.

    (1997)
  • D. Harris et al.

    Specificity of CEO human capital and compensation

    Strat. Manag. J.

    (1997)
  • Cited by (20)

    • Insider vs. outsider CEO and firm performance: Evidence from the Covid-19 pandemic

      2022, Finance Research Letters
      Citation Excerpt :

      In addition, outsider CEOs may have a mandate to lead and implement strategic changes as firms usually hire CEOs externally when there is strong demand for structural changes (Cannella Jr and Lubatkin, 1993). However, strategic changes led by outsider CEOs may disrupt performance (Jongjaroenkamol and Laux, 2017; Schepker et al., 2017; Zhang and Rajagopalan, 2010). Relative to insider CEOs, outsider CEOs often lack nuanced understanding of resources and constraints (Greiner et al., 2003) and the competitive environment (Castanias and Helfat, 1991), and therefore their strategic changes may fail to build upon existing competencies (Zhang and Rajagopalan, 2004).

    • CEO succession origin, audit report lag, and audit fees: Evidence from Iran

      2021, Journal of International Accounting, Auditing and Taxation
      Citation Excerpt :

      As CEO succession is particularly important to a firm’s strategy and successor, the board of directors is responsible for selecting the most competent candidate and ensuring a smooth leadership transition (Biggs, 2004). Jongjaroenkamol and Laux (2017) predict that a weak reporting system developed by the governance structure increases the likelihood that boards hire insider CEOs. However, over the past three decades, recruiting CEOs from outside the firm has become increasingly prevalent worldwide (Kaplan & Minton, 2012), particularly in Asian countries, such as China, Japan, and Korea (Liu, 2020), and Iran (Rezaee, Asiaei, & Safdel, 2021).

    • Is there a racial gap in CEO compensation?

      2021, Journal of Corporate Finance
      Citation Excerpt :

      While CEOs promoted from within used to be the norm in Corporate America, firms increasingly hire CEOs from the outside (Palomino and Peyrache, 2013). We therefore control for whether CEO is hired from outside the firm (external CEO; Jongjaroenkamol and Laux, 2017). We also control for firm characteristics that may influence CEO compensation.

    • The effect of financial reporting quality on CEO compensation structure: Evidence from accounting comparability

      2019, Journal of Accounting and Public Policy
      Citation Excerpt :

      Recently, Jongjaroenkamol and Laux (2017) proposed that CEOs’ incentive contracts can be affected by financial reporting that determines the quality of accounting information the board relies on in designing CEOs’ compensation schemes. Extending Jongjaroenkamol and Laux (2017), we argue that accounting comparability increases the usefulness of equity-based compensation by helping firms to detect CEOs’ misbehaviors with lower cost and, in turn, by reducing CEOs’ opportunistic incentives. Accordingly, we predict that firms are more likely to rely on equity-based compensation and increase pay-performance sensitivity as their financial statements become more comparable with those of their industry peers.

    View all citing articles on Scopus

    We thank S.P. Kothari (editor), an anonymous reviewer, Shane Dikolli, Ben Hermalin, Christian Laux, Paul Newman, and seminar participants at the University of Texas, University of Cambridge, University of California at San Diego, and the 2014 AAA Annual Meeting for many valuable comments and suggestions.

    View full text