Measuring reporting conservatism

The paper examines the power and reliability of the differential timeliness (DT) measure developed by Basu (1997) to gauge reporting conservatism. We identify certain characteristics of the information environment unrelated to conservatism that affect the DT measure and find that it is sensitive to the degree of uniformity in the content of the news during the examined period, the types of events occurring in the period, and firms' disclosure policies. Our tests, based on both actual and simulated data, indicate that assessing the extent of reporting conservatism using this measure requires the recognition of, and control for, these characteristics. We also find that the difference in the timeliness of reporting bad versus good news is likely to be more pronounced than previously reported. Further, we provide additional evidence on the negative association between the DT measure and alternative aspects of conservatism, suggesting that the exclusive reliance on any single measure to assess the overall conservatism of a reporting regime (firms, countries, or time periods) is likely to lead to incorrect inferences.