Improving profitability forecasts with information on earnings quality

Abstract

Prior literature documents the usefulness of DuPont disaggregation for many applications such as predicting firms’ future profitability, operating income, analysts’ forecast errors, and stock market returns. At the same token, researchers also emphasize the impact of earnings quality for economic decision models and the allocation of capital. However, there is a lack of understanding how earnings quality affects profitability forecast models. This paper comprehensively explores whether and to what extent different earnings quality factors moderate the accuracy of profitability forecasts that solely use financial statement information. I contribute to the existing literature along three dimensions. First, I reject findings from previous financial statement analysis literature that changes in firms’ profit margin do not provide incremental usefulness for predicting changes of future return on assets. After controlling for such influences, the incremental usefulness of this accounting signal increases significantly. Second, this paper contributes to the strand of earnings quality literature by suggesting an approach to include this set of information into forecasts. Hereto, I disentangle the two main drivers of earnings quality, fundamental performance and influences of the accounting system. Last, the paper adds to the literature on the perception of capital markets towards accounting information. By examining the awareness of financial analysts and investors of earnings quality information, I document that both sets of market participants seem to efficiently impound this information in their investment decisions.