Improvements in the Information Content of Nonfinancial Forward-Looking Performance Measures: A Taxonomy and Empirical Application

Abstract

In this paper, we study empirical refinements that increase the information content of nonfinancial performance measures (NFPMs) regarding future financial performance. We synthesize these refinements into a taxonomy that includes: (1) measurement, (2) timing, (3) interaction, (4) functional form, and (5) mediation. We classify recent work that uses these refinements to examine variations in NFPM information content. We also present an empirical application of the refinements taxonomy—using a NFPM previously shown to be valued by the market for Internet retailers, i.e., website stickiness—to further illustrate how theoretically driven choices of specific refinements can improve NFPM information content. The chosen refinements in this setting increase the adjusted R2 of our empirical specification, predicting one-quarter-ahead Return on Assets from 52 percent to a maximum of 78 percent. More important, these refinements reveal important new insights about website stickiness. In particular, website stickiness is a positive signal of future financial performance for firms with good websites, but a negative signal for firms with poor websites. Additionally, we find the magnitude of the positive and negative signals is asymmetric: the magnitude of the negative signal is significantly higher than that of the positive relation.