Perspectives on earnings quality : an integrated approach from european listed companies

Abstract

The quality of earnings is a summary metric in performance evaluation and a focal question to assess the quality of accounting information. A high-quality earnings number will reflect current operating performance, being a good indicator of future operating performance, and it accurately annuitizes the intrinsic value of the firm. The multidimensional nature of the earnings quality (EQ) concept has given form to a multiplicity of constructs and measures. The objective of the thesis is to provide a better and deeper understanding of the vectors of analysis in what concerns the dimensions of EQ concept, constructs and measures. We consider the multidimensional nature of the concept and highlight a “new” earnings quality perspective taking in account the virtuosities of the residual income model. It is proposed a empirical model which reinterprets rebuilding the linear information dynamics in relation to market value added and captures, in a composite measure, the tridimensional dimension of the EQ concept: persistence, predictability and informativeness of earnings. Our key findings are: - Imposing linear information structure, our proposed model provide a composite measure of EQ that captures the persistence, predictability and informativeness of earnings. Nonetheless, informativeness of earnings seems to capture per si all the relevant value information of earnings; - The valuation coefficient of net income differs from that of total accruals, and those of the four major accruals components differ from each other. These findings suggest that disaggregation of earnings into cash flow and total accruals, and total accruals into its major components aid in predicting market value added. - Predictions errors differ significantly when the linear information model (LIM) is imposed. - Our findings support the efficacy of drawing inferences from valuation equations based on residual income models that do not impose the structure implied by the model; - The magnitudes of the valuation parameter estimates and the values of adjusted R2 are better performed when we consider only positive earnings. So, it seems that loss cases have a dampening effect on the measures of the information content of earnings. They have a much weaker association with returns than profit cases. v Performing a separate industry estimation according to the system of equations for each earnings components (accruals and cash flows), we provide evidence that: - Informativeness of earnings is significantly higher in portfolios of industries with high earnings quality (high persistence of abnormal earnings and low (high) predictability of accruals (cash flows)) compared to portfolios of industries with low earnings quality (low persistence of abnormal earnings and high (low) predictability of accruals (cash flows)); - Explanatory power of earnings to explain market value added is significantly higher in portfolios of industries with high earnings quality (high persistence of abnormal earnings and low (high) predictability of accruals (cash flows)) compared to portfolios of industries with low earnings quality (low persistence of abnormal earnings and high (low) predictability of accruals (cash flows)). The results of the development of a measure instrument that allows to delimitate the basic constructs and measures of the EQ concept, through the application of an exploratory multivariate techniques analysis, namely, the factor analysis of principal components suggest six different dimensions of earnings quality: (1) time-series properties (persistence and predictability); (2) relevance; (3) accruals quality; (4) informativeness of earnings; (5) smothness and (6) timeliness. .