A Tale of Two Forecasts: An Analysis of Mandatory and Voluntary Effective Tax Rate Forecasts

Abstract

Disclosure theory predicts that the likelihood of voluntary disclosure increases with the level of noise in mandatory disclosure. We test this prediction by exploiting a unique setting where firms provide two forecasts of the same underlying metric – annual effective tax rates (ETRs). We find that managers are more likely to issue voluntary ETR forecasts when mandatory ETR forecasts contain more noise due to tax complexity. This finding suggests that managers resort to voluntary disclosure when mandatory disclosure constrains their ability to convey private information. Using analysts’ ETR forecast revisions to assess the informativeness of voluntary and mandatory ETR forecasts, we find that both forecasts are incrementally informative. However, analysts place a greater weight on voluntary ETR forecasts compared to mandatory ETR forecasts, especially when voluntary ETR forecasts are non-GAAP based and when discrete items are present. Overall, our study provides context specific evidence on the relation between voluntary and mandatory forward-looking disclosures, and on sophisticated financial statement users’ responses to both forms of disclosures.

Keywords: voluntary disclosure, mandatory disclosure, ETR forecast

JEL Classification: M40, M41