CEO's financial experience and earnings management

https://doi.org/10.1016/j.mulfin.2013.03.005Get rights and content

Highlights

  • We study the effect of CEO's financial experience on earnings management.

  • We distinguish between accrual-based earnings management and real earnings management.

  • The results show that CEOs with financial experience tend to do less real earnings management.

  • No evidence shows that CEOs with financial experience do either more or less accrual-based earnings management.

Abstract

We study whether Chinese CEOs with financial experience engage in more earnings management or less earnings management than those without such experience. In doing so, we distinguish between accrual-based earnings management and real earnings management. Overall, we find that CEOs with financial experience tend to do less real earnings management, while we find no evidence that they do either more or less accrual-based earnings management. Our findings tend to confirm that CEOs with financial experience provide more precise earnings information and higher quality financial statements.

Introduction

Although top managers, especially CEOs, are presumed to have a generalist's view, their past work experience shapes an orientation in their life, and this orientation can exert some influence on the firm's strategic choices (Hambrick and Mason, 1984). The impact of the CEO's work experience upon strategy adoption is well established. An early case study by Dearborn and Simon (1958) reports that when a group of executives from different functional areas are presented with the same problem, they define the problem largely in terms of goals and tasks in their respective functional areas, even though they are motivated to consider it from a company-wide viewpoint. Subsequent empirical studies provide further evidence. For example, Song (1982) finds that CEOs with operational work experience are more apt to favor internally developed diversification, while CEOs with experience in nonoperational functions are more inclined toward diversification through acquisition. Tyler and Steensma (1998) show that executives with technical work experience assess potential technological alliances more favorably than their counterparts with other work experience. Barker and Mueller (2002) claim that R&D spending is greater at firms where CEOs have career experience in R&D. All these studies show that the CEO's past functional experience significantly affects firm decisions.

In recent years, more and more companies have tended to hire CEOs with financial experience. Durfee (2005) reports that the percentage of Fortune 100 CEOs who were former CFOs increased from 12% to 20% over the past ten years. Cullinan and Roush (2011) find that out of 264 CEO appointments in publicly traded firms from 2001 to 2004, 15.48% of the CEOs appointed before SOX had accounting and/or finance experience, while 33.33% of CEOs appointed after SOX had it. This trend also holds in China. In 1995, only 0.9% of CEOs had financial experience, but the percentage increased to 5.71 in 2002, and remained above 5% from 2003 to 2010. If the CEO's past functional experience influences the firm's decisions, and CEOs with financial experience are more familiar with or expert in financial decisions, is there any difference in financial decisions between firms whose CEOs have had financial experience and firms whose CEOs have not? In this paper, we try to answer this question from the perspective of earnings management.

Many studies in the fields of finance and accounting focus on earnings management techniques, motivation, and governance mechanisms. For example, Baber et al. (1991), Bartov (1993), Roychowdhury (2006), and Beatriz and Encarna (2009) investigate why and how management manipulates earnings opportunistically. Klein (2002), Xie et al. (2003), and Bergstresser and Philippon (2006) study how earnings management can be effectively constrained through government mechanisms. Matsunaga and Yeung (2008) find that firms whose CEOs have financial experience provide more precise earnings guidance and improve the quality of financial disclosure, and they argue that the quality of a firm's financial disclosures is a function of the CEO's financial experience. To our knowledge, their article provides the first empirical evidence on the impact of CEOs with financial experience on earnings management.

However, there is still some room to study this topic. First, Matsunaga and Yeung (2008) used only one year (2004) of data to identify CEOs from the ExecuComp database, so their research employs a relatively small sample. Thus, it is important to understand whether the result is robust and broadly applicable. Furthermore, they examined the effect of the CEO's financial experience only on accrual-based earnings management through discretionary accruals, and did not consider earnings management through real activities.

Many studies demonstrate that earnings management includes both accrual-based and real activity earnings management (Schipper, 1989, Healy and Wahlen, 1999). Accrual-based earnings management (manipulation of earnings through exploitation of accounting discretion) is easy to detect, so it becomes less likely with the development of accounting standards and regulatory systems. As the manipulation of real transactions (cash flows from operations, discretionary expenses, and production costs) is harder to discern and more flexible, firms have switched to managing earnings in this way (Roychowdhury, 2006, Cohen et al., 2008). Bruns and Merchant (1990) show in their survey that financial executives indicate a greater willingness to manipulate earnings through real activities than through accruals. Through surveys and interviews with 401 financial executives, Graham et al. (2005) find that 78% of survey participants report that they would rather smooth earnings by real economic actions, while only 7.9% state that they would alter accounting assumptions to meet an earnings target. Thus we ask whether this choice is affected by the CEO's financial experience.

Executives with expertise and cognitive frameworks based partially on their career experiences are more likely to identify problems and seek additional information from the same functional domain and thus produce more successful decisions (Hambrick and Mason, 1984, Fredrickson, 1985, Walsh, 1988, Hitt and Tyler, 1991). CEOs with financial experience who consistently follow the accounting prudence concept have developed a steady work style. In addition, they are likelier to thoroughly understand the role of financial disclosures in reducing information asymmetry between corporations and investors (Matsunaga and Yeung, 2008) and helping market participants to assess corporate value (Hutton and Stocken, 2006). CEOs are responsible for earnings disclosure, and they must personally certify the accuracy and completeness of the financial information released by the company,1 both in the U.S and in China. Financial expertise and experience allow CEOs to better monitor firms’ accounting and disclosure policies. So compared to financially inexperienced counterparts, CEOs with financial experience provide more precise earnings guidance to analysts (Matsunaga and Yeung, 2008) and are less likely to manipulate earnings.

Using a somewhat broader definition of financial experience than do Matsunaga and Yeung, we define a CEO with financial experience as someone who has served as CFO, or in another high administrative position in finance or accounting, for example as chief accounting officer or vice-CEO for finance or accounting. In order to separate manager fixed effect from firm fixed effect,2 we begin our empirical investigation by identifying CEO turnover events of Chinese listed firms from 2002 to 2008. Using the difference-in-differences (DID) estimation method, and subgrouping this sample into pre- and post-turnover samples as well as treatment samples and control samples, we test how CEOs with financial experience affect accrual-based and real earnings management. While Matsunaga and Yeung (2008) find that firms with CEOs who have CFO experience tend to use less accrual-based earnings management, we find no relation between these variables. However, in accord with their arguments that the quality of a firm's financial disclosures is a function of the CEO's financial experience, we find robust evidence that CEOs with financial experience are less likely to manage earnings by real activities than CEOs without comparable experience, which shows that CEOs with financial experience provide more precise earnings information and improve the quality of financial statements.

Our paper provides the first empirical evidence that CEO's financial experience has impact on real earnings management, and advances the growing literature on determinants of earnings management. The paper also contributes to the literature in the effect of CEO's functional experience on firm's strategic choice. In addition, our findings partly suggest a possible explanation for the recent phenomenon that many firms hire CEOs with financial experience.

The remainder of the paper is organized as follows. Section 2 introduces the data, defines the main variables, and provides descriptive statistics. Section 3 presents empirical tests on whether having a CEO with financial experience affects earnings management, measured by discretionary accruals and by activities manipulation separately. Section 4 tests the robustness of the main results. Section 5 concludes.

Section snippets

Data and sample selection

We use data on Chinese listed firms from both the Shanghai and Shenzhen Stock Exchanges. CEO turnover information comes from the WIND database. CEO experience is identified by CEO resumes from the WIND database. Financial statement data come from the China Stock Market and Accounting Research database (CSMAR). We require each CEO turnover event in this test to have data available for at least two years in the pre-turnover period and two years in the post-turnover period (excluding the turnover

Difference-in-differences method

This paper, based on CEO turnovers, uses the difference-in-differences (DID) methodology (Imbens and Wooldridge, 2009), which exploits the richness of firm-level data while also avoiding the pitfalls of the regression-based method. The empirical model for the DID approach is as follows.|EMi,t|=α0+α1EERi*POSTt+α2EERi+α3POSTt+α4Xi,t+ϵi,t|DACi,t|=α0+α1EERi*POSTt+α2EERi+α3POSTt+α4Xi,t+ϵi,twhere i indexes firms, t indexes years, |EM| is the absolute value of real earnings management, and |DAC| is

Robustness tests

To test the reliability of our previous findings, we conduct the following robust checks:

Discussion

A growing literature in finance and accounting emphasizes the important role of reducing earnings management in improving corporate governance and investor protection. In China, the capital market has a relatively short history, with imperfect laws and regulations and weak investor protection. This leads to the failure of corporate governance mechanisms and to severe earnings management in listed companies. Meanwhile, compared to managers in countries with good corporate governance mechanisms

Conclusion

In light of the recent increase in the number of CEOs having financial experience, we examine whether that experience affects earnings management in Chinese A-share listed firms over the years 2002–2008. We find that the appointment of a new, financially experienced CEO lessens real earnings management. However, we find no evidence that the CEO's financial experience may affect accrual-based earnings management, which is inconsistent with the argument of Matsunaga and Yeung (2008), but is

References (43)

  • S. Roychowdhury

    Earnings management through real activities manipulation

    J. Account. Econ.

    (2006)
  • B. Xie et al.

    Earnings management and corporate governance: the role of the board and the audit committee

    J. Corp. Finance

    (2003)
  • C.P. Alderfer

    The invisible director on corporate boards

    Harv. Bus. Rev.

    (1986)
  • W.R. Baber et al.

    The effect of concern about reported income on discretionary spending decisions: the case of research and development

    Account. Rev.

    (1991)
  • B. Baik et al.

    CEO ability and management earnings forecasts

    Contemp. Account. Res.

    (2011)
  • V. Barker et al.

    CEO characteristics and firm R&D spending management

    Science

    (2002)
  • E. Bartov

    The timing of asset sales and earnings manipulation

    Account. Rev.

    (1993)
  • Beatriz, G.O., Encarna, G.S., 2009. Corporate Governance and Impression Management in Annual Press Releases,...
  • W. Bruns et al.

    The dangerous morality of managing earnings

    Manag. Account.

    (1990)
  • D.A. Cohen et al.

    Real and accrual-based earnings management in the pre-and post-Sarbanes Oxley periods

    Account. Rev.

    (2008)
  • D.C. Dearborn et al.

    Selective perceptions: a note on the departmental identification of executives

    Sociometry

    (1958)
  • Cited by (92)

    • The contagion effect of overconfidence in business group

      2024, International Review of Financial Analysis
    • Influence of earnings management on forecasting corporate failure

      2023, International Journal of Forecasting
      Citation Excerpt :

      Earnings manipulation can be addressed by detecting accounting abnormalities. Literature that documents such abnormalities has identified two earnings management tools: accruals and real activities manipulation (Campa & Camacho, 2015; Jiang et al., 2013). Because accruals relate directly to earnings, they are subject to being managed to alter earnings, and accordingly, they have been the focus of most empirical research on earnings management—as well as the main component managers use to alter reported earnings.

    • The effects of formal and informal CEO power on debt policy persistence

      2022, North American Journal of Economics and Finance
    View all citing articles on Scopus

    We thank an anonymous referee and Kenneth A. Kim for the helpful and excellent comments. We acknowledge the financial support from Program for New Century Excellent Talents in University (no. NCET-10-0796), the Fundamental Research Funds for the Central Universities and the Research Funds of Renmin University of China (nos. 11XNJ005, 11XNK029), and China National Natural Science Foundation (nos. 70972129 and 71172179). Any errors are entirely our own. Huang thanks the support of the Ministry of Education academic doctoral Prize in 2011.

    View full text