Thumbnail Image

Corporate Diversification, Earnings Management, and Financial Statement Readability

El Mouttaqui, Hajar
A Master of Science thesis in Finance by Hajar El Mouttaqui entitled, “Corporate Diversification, Earnings Management, and Financial Statement Readability”, submitted in November 2023. Thesis advisor is Dr. Feras Salama and thesis co-advisor is Dr. Kimberly Gleason. Soft copy is available (Thesis, Approval Signatures, Completion Certificate, and AUS Archives Consent Form).
I examine the implications of corporate diversification on the quality of financial reporting measured as earnings management and financial statement readability. The results reveal that diversified firms have a lower degree of earnings management than comparable portfolios of stand-alone firms and that the magnitude of reduction in earnings management increases with higher levels of diversification. Using path analysis, I find evidence that diversification decreases earnings management by reducing both earnings volatility and the need for external financing, and by increasing the propensity to pay dividends. The results are robust to using different measures of diversification and earnings management and continue to hold when I attempt to account for endogeneity using two different methods: the Heckman two-stage model and a propensity score matching analysis. To investigate the effect of diversification on financial statement readability, I use a widely utilized measure of financial statement readability, namely, the Bog Index. The analyses reveal interesting results. First, I document a negative association between diversification and financial statement readability, indicating that diversified firms have less readable annual reports. Second, the negative impact of diversification on readability is attenuated for firms with high ability managers, consistent with the notion that managerial ability is associated with better disclosure quality. These results are robust to using several measures of managerial ability and diversification. In addition, my results do not change when I estimate a Heckman two-stage model to account and correct for any protentional selection bias resulting from the decision to diversify.
External URI