تنوع و تعهد شرکت ها و مدیریت سود واقعی: یک رابطه غیر خطی Corporate diversification and accrual and real earnings management: a non-linear relationship

1. Introduction The evidence of earnings management practices is substantial (Dechow et al, 1998; Roychowdhury, 2006; among others). Executives may engage in earnings management to meet analysis expectations (Gunny, 2010), avoid losses (Burgstahler and Dichv, 1997; Roychowdhury, 2006), or maximize their personal benefits (Cheng and Warfield, 2005). Earnings management, though it may increase short-term earnings targets, is unlikely to increase long-term firm value (Alhadab et al., 2015; Roychowdhury, 2006). A recent line of research (Farooqi, Harris, and Ngo, 2014; Jirapon, Kim, and Mathur, 2008; Lim, Thong, and Ding, 2008; among others) has been conducted to examine the link between corporate diversification and earnings management. Our study extends that research strand in two important ways. First, these studies report important evidence on the difference in level of earnings management between diversified and non-diversified firms but are totally silent on the level of earnings management at different levels of diversification. If firms are different at each level of diversification in terms of organization complexity, strategic goals, financial resources, and investment opportunities (e.g., see Beattie, 1980; Gort, 1962; Wan, Hoskisson, Short, and Yiu, 2011; among many others), then managers’ motivations and abilities to use various techniques to manage earnings may be determined by level of corporate diversification. An illustration for this notion is the inverted U-shape relationship between degree of diversification and firms’ performance widely reported in the literature (Fryxell and Barton, 1990; Palich, Cardinal, and Miller, 2000; Rumelt, 1974; among others). We have developed a theoretical framework to investigate the hypothesis that earnings management may relate to level of corporate diversification in a non-linear pattern. Second, previous studies only investigate impact of corporate diversification on either accrual earnings management (Jirapon et al., 2008; Lim et al., 2008) or real earnings management (Farooqi et al., 2014). While Jirapon et al. (2008) and El Mehdi and Seboui (2011) find that corporate diversification reduces accrual earnings management, Farooqi et al. (2014) show higher levels of real earnings management in diversified firms. The findings suggest that firms may use accrual and real earnings management as substitutes for each other (Zang, 2012), though the evidence may not be sufficient to establish how corporate diversification impacts firms’ earning manipulation activities. Our second contribution is, that we simultaneously investigate the impact of corporate diversification on accrual and real activities earnings management and therefore are able to show novel evidence that managers of diversified firms use real and accrual earnings management as substitutes in managing earnings. Using a sample of US firms from 2001 to 2012, our paper focuses on investigating the effects of industrial diversification on earnings management as the theoretical arguments and empirical analysis in the diversification literature (Berger and Ofek, 1995; Lang and Stulz, 1994; Rajan, Servaes, and Zingales, 2000; among many others) are developed on firms’ industrial diversification (rather than geographical diversification).1