세무위험 및 조세회피의 재무보고의 불투명성과의 관계 (Tax Risk or Tax Avoidance and its Relation tothe Opacity of Financial Reporting)

서지정보

· 발행기관 : 한국세무학회
· 수록지 정보 : 세무학연구 / 36권 / 3호 / 9 ~ 54페이지
· 저자명 : 박종일, 김수인

초록

본 논문은 세무위험 및 조세회피가 재무보고의 불투명성과는 어떤 관련성이 있는지를 규명하는데 있다. 세무위험을 다룬 과거 연구는 조세회피와 같은 세무포지션이 세무불확실성을 증가시킬 것으로 보아(Dyreng et al. 2019) 조세회피처럼 세무위험 역시 정보의 불확실성을 높인다고 가정한다. 하지만 그동안의 세무위험을 다룬 연구들에서는 세무위험과 재무보고의 불투명성 간의 관계를 직접적으로 살펴본 연구는 전무하다. 따라서 본 연구는 세무위험(과거 5년간으로 측정된 GAAP 또는 Cash ETR의 표준편차)과 재무보고의 불투명성과의 관계를, 조세회피(과거 5년간의 장기유효세율)와 비교하여 살펴보았다. 본 연구는 재무보고의 불투명성 측정치로 DA(재량적 발생액), RM(실제 이익조정) 및 AQ(발생액의 질)의 다양한 측정치를, 또한 당기 외에도 장기 측정치를 이용하고, 2003년부터 2016년(2017년)까지의 상장기업을 분석하였다.
실증결과는 첫째, 세무위험은 재무보고의 불투명성(DA의 경우)에 대하여 대체로 음(－)의 관계를, 반면에 조세회피는 양(＋)의 관계였다. 둘째, 세무위험은 재무보고의 불투명성(RM의 경우)에 대해 명확한 관계가 나타나지 않은 반면에, 조세회피는 양(＋)의 관계였다. 그러나 세무위험을 연속변수 대신 지시변수로 측정하면 세무위험과 재무보고의 불투명성(DA, RM의 경우) 간에 음(－)의 관계로 나타났다. 셋째, AQ와 그 구성요소의 경우 세무위험은 InnateAQ(재량적 발생액의 질)와 양(＋)의 관계를, 조세회피는 AQ 및 InnateAQ에 대해 양(＋)의 관계였다. 마지막으로, 종속변수를 당기 대신 차기로 분석해도 앞서와 일치하였다.
이러한 본 연구결과는 조세회피는 재무보고의 불투명성과 전반적으로 양(＋)의 관련성이 있으나, 세무위험은 재무보고의 불투명성과 오히려 음(－)의 관련성으로 나타나 두 측정치 간에 재무보고의 질 측면에서 서로 다른 정보를 내포한다는 것을 보여주었다는데 의의가 있다.

영어초록

This study examines whether relation between tax risk/tax avoidance and opacity in financial reporting, through this we investigate tax risk and tax avoidance implications of financial reporting quality. In particular, this paper examine whether tax risk provides examples of how, and in which setting, information attributes about earnings quality, compared to tax avoidance. Prior research mainly focus on that tax avoidance is associated with aggressive financial reporting. For examples, Balakrishnan et al. (2012) and Park et al. (2017) find that firms’ tax aggressiveness is decreasing with financial reporting quality (i.e., accruals quality), whereas Shin et al. (2011), Kim et al. (2013), and Park and Park (2016) find the a negative relation between tax avoidance and discretionary accruals/real earnings management. Thus, there are competing views and mixed evidence on that as firms engage in greater levels of tax avoidance exhibit more or less financial reporting quality. However, what is unknown in the tax literature is whether firms with relatively high tax risk related to financial reporting quality. Therefore, we investigate whether either positive or negative the relation between tax risk and opaque financial reporting is an empirical question, which is a relatively unexplored area of tax research.
In this study, we use nine accounting based measures of opaque financial reports as the dependent variable － we use are：our first proxy for financial reporting opacity is based on firms’ discretionary accruals (hereafter DA), current DA, following Hutton et al. (2009) and Jeon and Park (2017), we use financial reporting opacity (hereafter OPACITY1, OPACITY2, respectively) in year t as the three－year (t－2, t) moving sum of the absolute value of annual DA, and the standard deviation of DA. We use the Dechow et al. (1995) and the Kothari et al. (2005) model to estimate the discretionary accruals. Our second proxy for financial reporting opacity is based on firms’ real earnings management (RM), following RM studies (e.g., Roychowdhury 2006), we consider three types of RM activities, by summing up the three individual measures (i.e., AbCFO, AbPROD, and AbDISE) as well as similar Hutton et al. (2009) and Jeon and Park (2017) methods, we construct two financial reporting opacity (hereafter OPACITY3, OPACITY4, respectively) in year t as the three－year (t－2, t) moving sum of the value of annual RM divided by 3, and the standard deviation of RM. Our third proxy for financial reporting opacity is accruals quality, following Francis et al. (2005), when we disaggregate the AQ into discretionary AQ versus innate AQ component (hereafter AQ, DiscAQ, and InnateAQ, respectively). In our set of tests, we examine the association between opaque financial reporting and tax risk/tax avoidance, as interest variable which we measure using the standard deviation of annual GAAP (Cash) ETRs over the five prior years (e.g., Guenther et al. 2017). Following prior ax avoidance studies (e.g., Dyreng et al. 2008), we calculate GAAP (Cash) ETRs over the period t－4 to t. For our empirical tests, sample firms are collected from the Korea Exchange from 1999 to 2017. We consider 6,880 firm－year observations for sample firm satisfying of the condition for firms with positive pre－tax income and ETR data from 2003 through 2016 (2017 based lag model) to construct two effective tax rate measures.
Our results are as follows. First, after controlling for various firm characteristics and tax avoidance that affect the dependent variable (e.g., opacity in financial reporting), we find a negative and significant relation between tax risk and opacity in financial reporting (i.e., DA, OPACITY1, OPACITY2). In contrast, we find no significant relation between tax risk and opacity in financial reporting (i.e., RM, OPACITY3, OPACITY4). However, tax risk are significantly and positively associated with innate AQ. Whereas, we find no significant association between tax risk and both AQ and discretionary AQ. Second, after controlling for various firm characteristics that affect the dependent variable, we find overall positive and significant relation between tax avoidance and opacity in financial reporting (i.e., DA, OPACITY1, OPACITY2, RM, OPACITY3, OPACITY4, AQ, InnateAQ). Third, when we set an indicator variable equal to 1 for firms with relatively high tax risk instead of continuous variable, we find the negative and significant relation between tax risk and opacity in financial reporting (i.e., DA, OPACITY1, OPACITY2, RM, OPACITY3, OPACITY4), and this relation is more pronounced. Lastly, we also utilize current as well as one－year－ahead dependent variable as measures of opaque financial reporting are similar results. Overall the results indicate that opaque financial reporting is significantly and negatively associated with tax risk. In contrast, the results for multivariate regressions consistently indicate that tax avoidance are significantly and positively associated with the measures of opaque financial reporting. However, where the InnateAQ are largely similar to those tax avoidance and tax risk.
In summary, these results show that the information attributes embedded in tax risk is associated with higher financial reporting quality, whereas tax avoidance is associated with lower financial reporting quality, thus both tax risk and tax avoidance are measures capturing different aspects of earning quality. The prior literature, however, emphasizes the informational properties of tax risk/tax avoidance about uncertain tax－related activities and does not investigate the implications of tax strategy on the overall corporate information quality. Our paper posits a much broader scope for the transparency implications of tax risk/tax avoidance firms, and we encourage future researchers to further explore the boundaries and implications of firms’ tax choices. Also, future research should focus on why tax risk and financial reporting quality are positively related and explore the firm characteristics. Our study is the first to examine the association between tax risk and financial reporting quality, thus contribution to the tax and accounting literatures.

참고자료

· 없음