세무보고 공격성이 재무분석가의 이익예측오차에 미치는 영향

This study investigates the relation between firm’s tax aggressiveness and quality of analyst forecasts measured as forecasting accuracy, bias, and dispersion among analysts. Prior research examining the effect of tax aggressiveness on firm value proxied by Tobin’s Q documents mixed results. Our study is distinct from prior research in that we focus on analysts who are more sophisticated than general investors and that we investigate how corporate tax aggressiveness affects quality of their forecasts. On one hand, corporate tax aggressiveness may increase firm value by reducing tax costs, thereby increasing net income and net cash flows. On the other hand, tax aggressiveness may exacerbate agency problem because managers can use tax aggressive activities to exploit their private benefits and such activities are difficult to monitor from outside(Slemrod, 2004; Desai and Dharmapala, 2006). Additionally, tax aggressive activities may involve severe monetary and non-monetary costs including reputation costs if tax authorities successfully prosecute any illegal activities(Slemrod, 2004). Further, prior research on tax avoidance argues that tax aggressive firms tend to produce less transparent financial reporting and that information asymmetry between managers and outside stakeholders are higher for such firms(Chen, Chen, and Shevlin, 2010; Kim, Li, and Zhang, 2011; Balakrishnan, Blouin, and Guay, 2012; Donohoe and Knechel, 2014; Kim and Cho, 2012). Empirically, Kim et al.(2011) document that corporate tax aggressiveness is positively related with stock price crash risk, Kim and Cho(2012) report a positive relation between tax aggressiveness and cost of debts, and Park and Chee(2016a) find a negative relation between tax aggressiveness and bond credit rating. In addition, Chen et al.(2010) and Balakrishnan et al.(2012) document a negative relation between tax aggressiveness and corporate transparency and Donohoe and Knechel(2014) document a positive relation between tax aggressiveness and audit fee reflecting higher audit risk for tax aggressive firms. Overall, these empirical findings suggest that corporate tax aggressiveness is associated with higher audit risk and higher financial reporting risk and the market requires risk premium for tax aggressive firms. Extending this line of research, we examine whether analysts produce lower quality forecasts for more tax aggressive firms. We construct proxy for tax aggressiveness using GAAP ETR　and CASH ETR introduced by Dyreng, Hanlon, and Maydew(2008) and CFO(cash flow from operations) ETR introduced by Guenther, Krull, and Williams(2014). Following Donohoe and Knechel(2014), we classify firms as tax aggressive if they are in the lowest quintile of GAAP ETR, CASH ETR, or CFO ETR by year within the same industry. Our sample covers KOSPI and KOSDAQ listed firms with available data in non-financial industries with fiscal year-end in December from 2004 to 2013. Our main findings are as follows. First, we find a positive relation between analyst forecast error and corporate tax aggressiveness. Specifically, more aggressive tax avoidance is associated with less accurate forecasts, more optimistically biased forecasts, and higher forecast dispersion, suggesting that tax aggressiveness increasing information uncertainty. Second, when we partition the sample into pre- and post-IFRS adoption period, we find that the positive relation between tax aggressiveness and analyst forecast error measured as accuracy, bias, and dispersion is stronger during pre-IFRS adoption period. This suggests that analyst forecast accuracy for tax aggressive firms improved after the adoption of IFRS. Lastly, the positive relation between corporate tax aggressiveness and analyst forecast error becomes less strong but still significant when we use ETR as a continuous variable to measure tax aggressiveness as commonly used by prior tax avoidance papers published in Korea. Overall, our results imply that the quality of analyst forecasts is lower for tax aggressive firms because analysts produce less accurate, more optimistically biased, and more dispersed forecasts for tax aggressive firms. There seldom exists research examining the relation between corporate tax aggressiveness and analyst forecast error using Korean data. Prior studies focus on the relation between tax aggressiveness and firm value and they document mixed results. Therefore, our research examining analyst perception about corporate tax aggressiveness contributes to the literature investigating the effect of tax aggressiveness on information uncertainty. Further,considering the importance of analyst role in decision making of investors and creditors on the capital market, our results provide implications to academicians, practitioners, and regulators.