Corporate social responsibility, real activities earnings management, and corporate governance: evidence from Korea

Abstract

We examine whether a firm’s corporate social responsibility (CSR) activities are associated with real activities earnings management (RAEM) based on stakeholder perspective. We also investigate whether corporate governance (CG) moderates the relationship between CSR and RAEM. Using a sample of 1432 firm-year observations of Korean-listed firms during 2005–2010, we find that socially responsible firms are significantly and negatively associated with RAEM. More importantly, we find that this relationship is moderated by CG as measured by a composite CG index. These results remain robust after we perform various sensitivity analyses. This study primarily contributes to the literature on CSR, CG, and RAEM by providing evidence for the moderating role of CG on the relationship between CSR and RAEM.

Keywords:

• accrual-based earnings management
• corporate governance
• corporate social responsibility
• real activities earnings management

JEL codes:

• G34
• M14
• M41

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. With respect to these accounting scandals, many scholars blame a decline in business morality (Coffee 2003).

2. In Korea, 46 firms published CSR report in 2011, but 96 firms published it in 2012, resulting in 109% sharp increase.

3. According to the 2013 United Nations Global Compact–Accenture CEO study on sustainability, 93% of the CEOs surveyed see sustainability or CSR as an important strategy to a firm’s success and 95% of the Global 250 firms issue sustainability reports (Accenture 2014).

4. Freeman (1984) suggests stakeholder theory arguing that CSR activities create more benefits than costs by keeping a good relationship with a firm’s related stakeholders such as customer, employee, community, and so on.

5. Graham, Harvey, and Rajgopal (2005, 32) survey and interview more than 400 executives and report the following:

“We find strong evidence that managers take real economic actions to maintain accounting appearances. In particular, 80% of survey participants report that they would decrease discretionary spending on R&D, advertising, and maintenance to meet an earnings target. More than half (55.3%) of the respondents state that they would delay starting a new project to meet an earnings target, even if such a delay entails a small sacrifice in value.”

6. Lim and Choi (2013) use 215 firm-year observations of Korean-listed firms during 2009–2011 whereas our study employs 1432 firm-year observations of Korean-listed firms during 2005–2010.

7. Leuz, Nanda, and Wysocki (2003) argue that EM practices vary across countries. Moreover, Reinhardt, Stavins, and Vietor (2008) report that different countries have different accounting statements, different levels of investor protection and legality of CSR. Gao (2011) argues that studies on CSR topic focus on the CSR practice in Western countries, while very little is known about the CSR practices in developing countries.

8. The stakeholder orientation is measured by the score of the principal factor of four alternative proxies for stakeholder orientation such as legal environment of a country in protecting labor rights, mandatory disclosure requirements on CSR issues, and two public awareness measures on CSR issues (Dhaliwal et al. 2012).

9. Analyst forecast errors are measured by the firm’s absolute value of average analyst forecast errors for forecasts made in year t for the earnings in year t (Dhaliwal et al. 2012).

10. Beurden and Gössling (2008, 407) suggest that “there is indeed clear empirical evidence for a positive correlation between CSR and financial performance.” Peloza (2009) shows that 68, 15, and 22% articles published during 1972–2008 are positive, negative, and mixed relationship, respectively.

11. Freeman (1984, 46) defines stakeholders as “any group or individual who can affect, or is affected by, the achievement of a corporation’s purpose. Stakeholders include employees, customers, suppliers, stockholders, banks, environmentalists, government, and other groups who can help or hurt the corporation.”

12. ABEM is known to affect firm value in a minimal or no level (Subramanyam 1996).

13. Despite the negative effect of RAEM on firm value, why do managers prefer RAEM to ABEM? First, ABEM is more subject to the risk of regulators’ monitoring and investors’ litigation than RAEM (Cohen, Dey, and Lys 2008). Second, ABEM in general occurs after the end of the fiscal year because managers should consider the auditor’s position to avoid the audit risk that they may not receive an unqualified opinion from the external auditor (Zang 2012). Third, RAEM is likely more difficult to detect than ABEM (Cohen, Dey, and Lys 2008).

14. With respect to CSR–ABEM studies, there are several studies. Some studies support the contention that CSR constrains ABEM (Choi and Pae 2011; Scholtens and Kang 2012). However, other studies argue that a positive relationship exists between CSR and ABEM (Prior, Surroca, and Tribó 2008; Gargouri, Shabou, and Francoeur 2010). Moreover, Chih, Shen, and Kang (2008) show a mixed result using 1653 observations in 46 countries. Taken together, despite mixed results, a majority of prior studies suggest that CSR reduces ABEM.

15. We complement the study of Kim, Park, and Wier (2012) by expanding their research to the Korean setting and using a different CSR index, and the study of Lim and Choi (2013) by using extended sample period and larger sample size.

16. A few studies using a composite CG index also show mixed results. For example, Jiang, Lee, and Anandarajan (2008) document that CG is negatively and significantly associated with both ABEM and a small earnings surprise. However, Larcker, Richardson, and Tuna (2007) argue that CG has a mixed association with ABEM and little relation to accounting restatements.

17. KOSPI stands for the Korea Composite Stock Price Index.

18. KOSDAQ stands for the Korea Securities Dealer Automated Quotations.

19. We are aware that KEJI index includes CG components like KLD index does. However, we initially use total score without any adjustment due to following reasons. First, several studies (e.g. Choi, Kwak, and Choe 2010; Oh, Chang, and Martynov 2011) using KEJI index employ total CSR score without CG adjustment. Second, we think that partial CG components included in KEJI CSR score would not significantly influence our empirical results because KEJI CSR and CG scores are measured by mutually independent and authoritative organizations in Korea. However, adjusted CSR score excluding CG components will be employed to test robustness of our results in Section 4.3.3.

20. CG dummy variable (high CG score group = 1 and low CG score group = 0) may help mitigate the problem of measurement errors in raw CG score. Thus, we additionally test CG dummy variable instead of GOV_SCORE to confirm whether our results remain unchanged. We find that the untabulated results remain robust.

21. Based on the sample distribution by the two-digit KSIC code industry, the high CSR score industries are pharmaceuticals, medicinal chemicals, and botanical products (48.07 points), chemicals and chemical products (46.37 points), and electronic components, computer, radio, television & communication equipment products (46.31 points), whereas the low CSR score industries are publishing activities (42.26 points) and pulp, paper, and paper products (43.11 points). This distribution indicates that industries that need to prevent environmental pollution like chemicals or enhance firm’s brands and images like medicine or electronic products are more likely to proactively engage in CSR activities to meet stakeholders’ expectation.

22. We gratefully acknowledge insights provided by a reviewer.

23. In Korea, foreign investors from North America and Europe countries hold a significant portion of equity shares of Korean firms. Thus, these foreign investors tend to exert their power on Korean firms to establish transparent CG and drive them to actively engage in CSR (Oh, Chang, and Martynov 2011).

24. There are several reasons why CSR score is different from CG score even though the correlation between two scores is somewhat high. First, CSR score and CG score have different perspectives in that CSR score emphasizes diverse stakeholder perspective such as communities, customers, and employees as well as shareholders, whereas CG score mainly stresses the protection of shareholders. Second, the individual components of CSR score index are different from those of CG score index although some of the elements seem to be overlapping.

25. Kim, Park, and Wier (2012) also construct CSR scores excluding CG items to disentangle the similar effect of CSR and CG.