Family firms, tax avoidance, and socioemotional wealth: evidence from tax reform in Taiwan

Abstract

In 2018, Taiwan implemented a tax reform package that abolished the imputation regime, providing managers stronger incentive to engage in tax avoidance. This paper employs this unique setting to examine how the Taiwan’s 2018 tax reform affects the relation between family firms and tax avoidance based on a socioemotional wealth (SEW) view. I find that family firms on average exhibit lower tax aggressiveness than nonfamily firms. Further, I discover that family firms are disinclined to engage in tax avoidance to a greater extent after the policy change. Moreover, I determine that the negative association between family firms and tax avoidance is more salient for firms with strong control-enhancing mechanisms. Further, additional analyses indicate that family firms may also perceive tax avoidance to be potentially risky and value-destructive and thus are disinclined to engage in it. Collectively, my evidence is consistent with the predictions under the SEW view.

Appendices

Appendix 1: Test of tax avoidance before and after the 2018 tax reform

To test how the 2018 tax reform in Taiwan affects tax avoidance activities, I estimate Eq. (5) as follows:

$$TA_{{{\text{it}}}} = \beta_{0} + \, \beta_{{1}} Post_{{{\text{it}}}} + \, \beta_{{{2}\sim {\text{k}}}} X_{{{\text{it}}}} + YearDummies + IndustryDummies + \, \varepsilon_{{{\text{it}}}}$$

(5)

Post is a dummy, equal to 1 if year is 2018 or 2019 and 0 otherwise. X represent a set of time-variant firm-year control variables, which are the same as the controls in Eq. (1) and defined in “Appendix 2”. Results from model (5) using ordinary least squares (OLS) regressions with standard errors adjusted for heteroskedasticity and within firm clustering are presented in Table 12.

Table 12 Results from regressions of tax avoidance on a dummy variable, post, and other controls

Column (1) of Table 12 illustrates that coefficient on Post is 0.009 (t = 4.29). Further, I rerun the test using TA_gaap instead of TA_cash as the dependent variable. Column (2) reveals that the coefficient on Post is 0.008 (t = 4.21). Both are significant at the 1% level. These results suggest that tax avoidance increases after the elimination of imputation system before controlling for factors affecting tax avoidance activities.

Further, after controlling for all other factors, I find that the coefficient on Post is 0.012 (t = 4.44), which is significant at the 1% level and suggests that tax avoidance significantly increases after the 2018 tax reform. Finally, I rerun the test using TA_gaap instead of TA_cash and report the results in column (4). The coefficient on Post is 0.008 (t = 4.21). The finding again indicates that tax avoidance activities become significantly more active after the 2018 tax reform.

Taken together, these results are consistent with McClure et al. (2018) and Amiram et al. (2019) and support the position that the elimination of imputation provides managers with stronger incentive to engage in tax avoidance.

Appendix 2

See Table 13.

Table 13 Definitions of key and control variables