4.2.4. Results Hypothesis 5—Board Activity

The results of the panel regression suggest a positive, though insignificant relation between board activity and earnings management. This implies that board meeting frequency does not seem to have a direct effect on earnings management, in contradiction to what was expected in the hypothesis and the results of Vafeas (1999), Xie et al. (2003) and Daghsni et al. (2016). The result is however in line with previous studies conducted by Ahmed (2007) and Ahmed (2007). It is worth noticing that the *p*-value of 0.103 is close to a 10 percent significant level.

#### 4.2.5. Results Hypothesis 6—Audit Committee

Further, the regression analysis points out that an audit committee who supervises the financial reporting and disclosure negatively affects the occurrence of earnings management. This is in line with the hypothesis and the studies conducted by Klein (2002) and Dechow et al. (1996). The finding implies that the audit committee's role in board matters contributes to create trust by securing internal control of financial reporting and that the firm complies with laws and regulations. In addition, one could argue that the regular contact they have with the firm's external auditor could be effective in reducing agency conflicts as they weigh divergent views to produce a more balanced and accurate financial report.

Finally, the control variables behave as expected and are consistent with other earnings management studies (Iqbal et al. 2015; Daghsni et al. 2016). Firm size is found to be negatively related with earnings management, indicating that the occurrence of earnings management is decreasing in line with the size of the firm. The results further show that ROA and ROE negatively affects earnings management, suggesting that earnings management decreases as firm performance and profitability increases. In addition, all control variables are significant.

## *4.3. More Discussion*

We do acknowledge the potential of endogeneity issues in our analysis, as e. g. omitted variables. We are also aware of the important role of firm size in this kind of research, and thus can affect the independent and dependent variables simultaneously—see Coles and Li (2020) for a comprehensive discussion. Moreover, we observe that robustness tests can weaken our findings to some degree, however our main message of the analysis remains.

#### **5. Conclusions**

Cited as the next supermodel for corporate governance (The Economist 2013), it is of interest to examine corporate governance practices within the Nordic model of corporate governance. The purpose of this study was to provide insight to better assess the relation between Nordic corporate governance practices and earnings management, and potentially highlight the benefits of the model. The robust multivariate regression analysis under the fixed effect estimator has been used for estimation, while the absolute value of discretionary accruals is used as a proxy for earnings management (Hribar and Nichols 2007).

The presence of employee representation on the board and the presence of an audit committee are both practices that seem to reduce the occurrence of earnings management. The negative relation between the presence of an audit committee and earnings management is already well-established in the earnings management literature (Klein 2002; Dechow et al. 1996), while the findings of employee representation is to some extent new insight. Our findings may suggest that employee representatives provide a credible channel for information, contributing to a richer information environment. This can mitigate agency costs and earnings management and could imply that there are other important aspects of independence that should be taken into consideration to improve the quality of the directors. As for the results regarding share ownership by directors, the findings indicate that large proportions of minority shareholders on the board could give the directors incentives to pursue higher-risk strategies to generate larger financial returns. Finally, board activity and directors as majority shareholders both presented insignificant relations to earnings management. Still, their implications on earnings management may be of interest.

The contribution of this study is not without limitations. First, by using discretionary accruals as a measurement for earnings management one relies solely on proxy measures. Hence, one cannot exclude the possibility that the findings are subject to more natural accounting explanations than earnings management. Second, the relatively small sample size could affect the accuracy of the estimations. Third, our model is not without econometric challenges, and, finally, the corporate governance model may not be enough in capturing the omission of other corporate governance variables. These limitations may constrain the validity of the findings.

**Author Contributions:** Conceptualization, F.K., A.S. and A.V.; methodology, A.T.H.; software, A.T.H.; formal analysis, A.S. and A.V.; investigation, A.S. and A.V.; data curation, A.S. and A.V.; writing—original draft preparation, A.S. and A.V.; writing—review and editing, F.K.; supervision, F.K.; project administration, F.K. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Acknowledgments:** We appreciate the thoughtful comments and constructive suggestions from two anonymous reviewers.

**Conflicts of Interest:** The authors declare no conflict of interest.
