We revisit the asymmetric timeliness of earnings as proposed by Basu (1997). For a large sample of US firm years from 1970-2019, we show that evidence of asymmetrical timeliness with respect to bad economic news is predominantly, if not entirely, restricted to the idiosyncratic component of earnings. We further show that this result is robust to controlling for several identified forms of bias (Badia et al. 2021) and possible asset curtailments (Lawrence et al. 2018), as well as the use of lagged earnings changes as an alternative method for estimating asymmetric timeliness. Our results support the argument in Watts (2003) that asymmetric timeliness has evolved primarily as a response to information asymmetry issues in contracts that rely on accounting information.