This study uses a unique institutional setting in China to investigate empirically the association between the organizational form of CPA firms (unlimited liability versus limited liability) and the reporting conservatism of auditors. Based on a sample of 5,007 audits of Chinese listed companies from the period of 2000 to 2004, we find that auditors in unlimited liability partnership firms are more likely to issue modified audit opinions than are auditors in limited liability CPA firms. This auditor conservatism stems from distressed firms with going-concern opinions, and we find no statistical evidence that partnerships give more modified opinions for non-distressed firms. In addition, auditors are less likely to issue modified reports after they incorporated the firm in the limited liability form. These analyses support our hypothesis that a limited liability regime induces lower auditor reporting conservatism. Our study contributes to the broader debate on liability reform in the auditing profession.

We examine how two attributes of preventive and detective controls affect employee performance and employee motivation. Specifically, we examine how the extent to which controls (1) restrict employees’ autonomy, and (2) provide more or less timely feedback impacts employees’ performance and intrinsic motivation. These characteristics are the defining differences between preventive and detective controls in that preventive controls restrict employee autonomy relative to detective controls and preventive controls always provide immediate feedback; whereas, detective controls can provide either immediate feedback or delayed feedback. We conduct an experiment to examine how and why these two types of formal controls impact employees’ performance in an incomplete contract setting in which one dimension of the employees’ task is compensated and one dimension is controlled. The results show that detective controls with more timely feedback improve employees’ performance toward the control objective, without affecting their intrinsic motivation. In contrast, the restriction of autonomy associated with preventive controls, has no additional effect on employees’ performance toward the control objective over detective controls with timely feedback but significantly reduces employees’ motivation. Neither control characteristic has a significant effect on employees’ performance on the compensated dimension of the task, suggesting that monetary incentives continue to provide an effective motivation. Our results reveal the importance of designing and implementing controls that provide timely feedback but do not restrict autonomy.