An Empirical Investigation of Non-GAAP Exclusion Quality Indicators

Abstract

We examine commonly used indicators of low non-GAAP earnings quality and find that, contrary
to researchers’ intentions, most indicators identify firm-quarters with excluded earnings
components that (1) have relatively low persistence for future firm performance, and (2) improve
the value relevance of earnings when excluded. We conclude that existing indicators of non-GAAP
earnings quality perform poorly at identifying low-quality non-GAAP earnings. We propose a new
indicator that relies on identifying firm-quarters where GAAP earnings quality is high but
managers disclose non-GAAP earnings anyway. We find this indicator performs far better at
identifying firm-quarters with low-quality non-GAAP earnings. Our results are robust to a variety
of specification checks.

Keywords: Non-GAAP Earnings, Voluntary Disclosure, Earnings Quality, Benchmark Beating

JEL Classification: M40, M41