No. 81: Does the story matter? Putting financial statement numbers into context using XBRL data

Abstract

Using XBRL data in a tax disclosure setting, we create a new measure of “discussed numbers” that captures the context of firms’ financial reporting—namely the qualitative discussion of numbers in financial statements. On the determinants side, we show that our measure is indicative of managers explaining higher GAAP effective tax rates (ETRs) and greater volatility in both GAAP and cash ETRs. Our measure also increases in geographic complexity, decreases in age, and is associated with concise yet complex narrative. On the consequences side, we find that discussing tax numbers improves analyst forecast accuracy of implied ETRs and reduces the dispersion of those forecasts. We confirm that the benefits to analyst forecast properties manifest in sentences that explain tabulated numbers in financial statements, and not only in sentences that simply contain or repeat numbers. Finally, exploiting the richness of the tax setting, we find that greater discussion of complex tax numbers, including FIN 48 unrecognized tax benefits, deferred taxes, and the valuation allowance, are especially helpful to analysts in forecasting the ETR. Overall, this study enhances our understanding of corporate disclosure practices by showing that the interrelation of numerical values and narrative discussion, especially in a complex setting, can improve the information environment for financial statement users.