Using a sample of 8,788 observations from Chinese listed firms from 2001 to 2009, we investigate the difference in the information content of discretionary accruals between those audited by Top 10 and non-Top 10 accounting firms. We find that discretionary accruals are positively related to the probability of informed trading (PIN) and bid-ask spreads; the larger the discretionary accruals, the ... [Show full abstract] higher will be PIN, the level of information asymmetry, and bid-ask spreads. Furthermore, we argue that the significance level of the correlation between discretionary accruals and PIN/bid-ask spreads differs between companies audited by Top 10 and by other firms. The positive relationship between discretionary accruals and PIN/bid-ask spreads is significantly weaker in companies audited by the Top 10. This indicates that the Top 10 firms are better at restraining opportunistic earnings management, and also that the information content of their clients’ discretionary accruals is higher and their information asymmetry lower. Our study contributes to a better understanding of the two types of earnings management through manipulating discretionary accruals; one is based on delivering value-related information and the other arises from opportunism.