Does Earnings Management Influence Credit Rating Changes in Subsequent Periods?: An Analysis of KRX Firms

There is a potential for firms to engage in earnings management to influence a credit change (Ali and Zhang 2008; Jung et al. 2013; Alissa et al. 2013). However, Mali and Lim (2016) find that firms that engage in earnings management in period t do not experience a credit rating increase in period t+1; rather these firms are more likely to experience a decrease. Their evidence suggests that credit rating agencies capture both accrual earnings management (AEM) and real earnings management (REM) in the credit watch period (t-1 to t). Firms that potentially experience a credit rating change must experience a 1 year credit watch period in advance of potential change. Therefore, we conjecture a firm may engage in REM in period t-2 because firms are under less scrutiny from rating agencies, and REM is more difficult to detect. We use the residual from the Dechow et al. model (1995) and Kothari’s (2005) model as proxies for AEM. We use models suggested by Cohen et al. (2010) as proxies for REM. Using a sample of 1,481 observations from 2002 to 2013, we find a negative association between AEM in period t-1 and credit rating changes in period t, suggesting that credit rating agencies consider low levels of abnormal accruals as a form of strong corporate governance. We find a positive association between REM in period t-2 and a positive change in period t, suggesting that a firm may use earnings management to influence credit ratings increase. Moreover, we find no association between credit rating decreases and earnings management suggesting that firms have the opportunity to engage in earnings management to influence credit ratings. Overall, the results suggest that firms may engineer their financial structure by using REM in period t-2 to increase the probability of credit rating change.