배당이 미래 이익의 질을 전달하는가?

We examine whether dividend paying status provides information about earnings quality. Specifically, in this paper, we investigate whether dividends are informative about future earnings quality. There is a long line of literature that investigates whether managers use dividends to signal the future prospects of their firm - known as the dividend information content hypothesis. According to the information content of dividends hypothesis, managers’ dividend decisions provide information about the earnings prospects of their firms. Under the most common interpretation of this hypothesis, changes in firms’ dividends map directly into changes in future earnings. However, that idea is not strongly supported in the data. On the other hand, recent studies indicate dividends are associated with higher-quality earnings (e.g., Hanlon et al., 2007; Skinner and Soltes, 2011; Tong and Miao, 2011; Caskey and Hanlon, 2013; Kim et al., 2016 etc.). Our study extends the previous literature by examining whether dividends’ information is associated with earnings quality. In other words, in this paper, we examine whether dividend paying firms have better earnings quality about future than non-dividend paying firms. Specifically, we analyze managers use both discretionary accruals and real earnings management activities, these two methods to achieve their opportunistic goals as well as accruals quality as a proxy for the state of financial reporting quality. Our study differs from the prior research as their focus is on whether dividend paying status are informative about current earnings quality, while our focus is on whether dividend paying status is informative about current earnings quality. Meanwhile, prior research relies on various empirical measures for earnings quality as no one single measure exists (Dechow et al., 2010). Thus, we utilize three empirical proxies in order to corroborate and triangulate the inferences that can be drawn on the association between dividends and earnings quality. The first proxy is discretionary accruals (hereafter DA) calculated using both Dechow et al. (1995) and Kothari et al. (2005) models. The second proxy is real earnings management (hereafter RM) calculated using the Roychowdhury (2006) models. The third proxies are based on the mapping of accruals onto lagged, current, and future cash flows using both Dechow and Dichev (2002) and Francis et al. (2005) models of accrual quality (hereafter AQ). For analysis, similar to following Hutton et al. (2009), Jeon and Park (2017), and Francis et al. (2005), this paper measures about future earnings quality as the subsequent three years (t= 0 to +2) standard deviation of DA, RM, and AQ, respectively. Our sample covers KOSPI and KOSDAQ listed firms in non-financial industries from 2002 to 2014 (based on the dependent variable from 2004 to 2016). The empirical findings of this paper are following. First, after controlling for several factors that affect earnings quality, we find that the coefficient on the dummy variable for dividend paying status is significantly negative for all three about future earnings quality proxies. That is, compared to non-dividend payers, dividend payers have lower standard deviation of future earnings quality measures based on DA, RM and AQ, respectively. Thus, this implies that dividend paying firms’ future earnings quality are higher than non-dividend paying firms’, this results provides support for the dividend information content hypothesis. Second, we use an alternative definition of the dividend indicator variable to identify persistent dividend paying firms (e.g., if the firm pays dividend for five contiguous years from year t-4 to year t). we find that compared to firms that do not pay dividends or pay dividends for less than five contiguous years, persistent dividend paying firms that have lower standard deviation of future earnings quality measures based on DA, RM and AQ, respectively. This results show that relatively persistent dividend paying firms having a significantly higher earnings quality about future do not pay dividends or/and pay dividends for less than five contiguous years. Finally, when we partition the samples into KOSPI and KOSDAQ listed firms, our main findings regarding the effect of dividend paying status on the future earnings quality are robust to stock exchange type of listed firms. Overall, this results evidence supports the idea that dividends are informative about future earnings quality. In that sense, we provide new evidence on the dividend information content hypothesis by investigating the association between dividend paying and future-year earnings quality for firms that pay dividends in the current year as compared to firms that do not pay a dividend. As a result, we conclude that dividend paying status contains additional information about future earnings quality that helps investors better forecast future earnings. That is, investors are better able to predict future earnings quality for dividend paying firms compared to non-dividend paying firms. And the findings of this paper also contribute to the related literature on the positive association between dividend and earnings quality by providing empirical results. Furthermore, the findings of this paper also provide an important implication and valuable insights for accounting researchers, practitioners, and regulators who are interested in dividend paying status per se is indicative of firms’ future earnings quality.