We document 45 specific allegations related to audit deficiencies based on GAAS, as detailed in 141 AAERs and 153 securities class action lawsuits over the violation years 1978–2016. Next, we use these allegations to validate popular proxies of audit quality. Of all the audit quality proxies, we find that restatements consistently predict all of the top six most cited audit deficiencies. The ratio of audit fees to total fees and the presence of a city specialist auditor predict five of the most cited deficiencies. Overall, our results suggest that the predictive power of audit quality proxies depends on (i) the settings that researchers are interested in and (ii) the specific audit deficiencies hypothesized to matter in the investigated setting. For instance, future studies related to auditor independence might consider using restatements and the ratio of audit fees to total fees as proxies of audit quality.

我们记录了基于GAAS的45项与审计缺陷相关的具体指控，详细说明了1978-2016年违规期间的141项AAER和153项证券集体诉讼。接下来，我们使用这些指控来验证流行的审计质量代理。在所有审计质量代理中，我们发现重述一致地预测了所有被引用最多的前六项审计缺陷。审计费用与总费用的比率以及一名城市专职审计员的存在预示了五个被引用最多的缺陷。总体而言，我们的结果表明，审计质量代理的预测能力取决于（i）研究人员感兴趣的环境和（ii）假设与调查环境相关的特定审计缺陷。例如，