The effect of management style on financial statement comparability: evidence from Korean business groups

ABSTRACT

In this study, we investigate whether firm-pairs in the same business group have more comparable earnings than firm-pairs in different business groups. We find evidence consistent with our hypothesis that accounting earnings comparability is greater when two firms are members of the same business group. Additionally, we find that accounting earnings comparability for firm-pairs in the same business group is greater where insider ownership of business groups is higher and the exchange of board personnel is more frequent. In sum, we confirm that ‘management style’ in each business group contributes to an improvement in accounting earnings comparability across member firms in the same business group.

KEYWORDS:

• Business group
• comparability
• management style
• accruals
• insider ownership

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. ‘Tunneling’ and ‘propping’ refer to the transfer of wealth, either from minority shareholders to controlling shareholders via ‘tunneling’ or from healthier to more financially troubled business groups via ‘propping’ (Bae, Cheon, and Kang 2008).

2. A ‘co-insurance effect’ implies that two or more firms whose earnings streams are uncorrelated or less than perfectly correlated, reduce their default risk, thereby increasing their debt capacity (Lewellen 1971).

3. For example, founder CEOs view their role as maintaining existing strategies and values of a firm, while professional CEOs view their role as bringing about change in their organization and values of shareholders (Mullins and Schoar 2016).

4. Bertrand and Schoar (2003) show that ‘management style’ affects management decisions such as acquisitions and cash holdings. Malmendier and Tate (2008) also show that CEOs who win awards (e.g. Business Week’s ‘best managers’) underperform relative to their peers after winning the award.

5. For example, managers promoted from legal career tracks tend to guide expectations down due to sensitivity to litigation risk, and those promoted from accounting or finance career tracks tend to disclose less frequently and more precisely due to sensitivity to underestimating upcoming earnings (Bamber, Jiang, and Wang 2010).

6. De Franco, Kothari, and Verdi (2011) suggest financial statement comparability based on the mapping of earnings to stock returns across firms in addition to an approach based on the covariation in earnings across firms. Because most business groups also have in-house rules that affect member firms’ choice in accounting alternatives, the earnings covariation metric used in Francis, Pinnuck, and Watanabe (2014) is a far more appropriate proxy for testing our hypothesis directly. De Franco, Kothari, and Verdi (2011) and Francis, Pinnuck, and Watanabe (2014) use adjusted $R^2$ for accounting comparability, so we tested it again using an adjusted $R^2$ and the results are similar to those in Tables 4–6.

7. Francis, Pinnuck, and Watanabe (2014) use firm-pairs with non-overlapping four-year periods for earnings covariation, to avoid concerns over the non-independence of error terms. Easton and Pae (2004) conduct an annual regression because omitted variable bias could affect the results (e.g. accounting rules). Considering the short study period, we therefore conduct an annual regression for earnings covariation to mitigate these concerns, and find that the coefficient SameBG of the weighted average number of firm-pairs is significantly positive at the p < 0.10 level (two-tailed).

8. The coefficient Same Big 4 is 0.001 in the total accruals-difference test, and the coefficient Same Big 4 is 0.002 in the earnings covariance test in Francis, Pinnuck, and Watanabe (2014). It is not possible to compare these directly, but the coefficient $SameBG$ in our study is somewhat higher than the coefficient Same Big 4 in Francis, Pinnuck, and Watanabe (2014).

9. Bae, Kang, and Kim (2002) argue that controlling shareholders’ small ownership causes agency problems between controlling and minority shareholders. Therefore, we expect that the substantial power of controlling shareholders is greater when their ownership is higher because there are no or few agency problems between controlling and minority shareholders.

10. We exclude 25 M&A firms and 964 firm-pairs from the same business group via the search terms ‘Merge’ or ‘Acquisition’ in Google.

Additional information

Funding

We sincerely thank the anonymous reviewers for the helpful comments and this research was financially supported by Hansung University.