績效審計有效性之影響因素

績效審計因應新公共管理而發展，作為政府績效管理的重要工具，理論上應能藉由審核意見，促使受查機關對於後續施政方針或業務流程做出改善，並提升政府整體公共課責，然而，根據國外推行績效審計多年經驗，績效審計制度的建立與績效審計是否發揮效用，並不全然存在因果關係，甚至有研究發現績效審計執行成效其實未如預期。審計部自2009年開始積極推動績效審計，至今已13年，本文欲探討我國績效審計與政府施政效率之間的關係，以及影響績效審計能否發揮效果的各項因素。過去有關績效審計之實證研究，分別從「增進財務效益」、「受查機關人員對於績效審計有效性的感知程度」、「受查機關對於審核意見的回應及改善情形」及「績效審計引發的公開辯論或輿論關注」等四個面向切入，檢驗績效審計的執行成效，並證實「審核意見品質」、「審核意見缺失類型」、「外部使用者對於審核報告的使用程度」、「受查機關對於績效審計的態度」、「查核頻率」、「受查機關層級」、「審核意見與受查機關政策方向的一致性」等七項因素，的確會影響績效審計的有效性。本文運用審計部在2014至2019年間，查核原住民族委員會、環境保護署、海洋委員會的審核意見及受查機關聲復內容，進行檔案研究，發現高達九成審核意見均能獲得受查機關表達同意，且受查機關人員也大多認同審核意見內容，我國績效審計應已發揮相當程度的效果，然而受查機關提出具體改善措施的比率卻下降至六成七，顯見即使受查機關人員認為績效審計是有效的，也不必然會根據審計人員提出的審核意見作出改變。另根據卡方檢定分析結果，不同的審計類型及審核意見缺失類別，確實對於受查機關研提改善措施與否有顯著影響，兼具遵循審計及績效審計兩種性質的審核意見，受查機關確實較可能研提具體的改善措施；審核意見缺失類別方面，涉及「預算或人力不足」及「監督管考機制未完善」的審核意見，有助於受查機關爭取相關人力預算或管考受補助地方政府，受查機關較可能研提改善措施，而涉及「法令規範尚未完備」的審核意見，則受查機關研提改善措施的可能性較低。最後，根據訪談內容，審計人員及受查機關人員均認為追蹤覆核審計確實會帶給受查機關不小壓力，有助於促使受查機關加速改善，然而，訪談內容並未反映在實際的聲復內容上。本文發現追蹤覆核審計經常著重在查核議題的重要性及外界關注程度，而非受查機關的「待改善程度」，即使受查機關已經在上一年度的聲復內容提出具體改善措施，審計人員仍可能礙於各界壓力或是總決算審核報告揭露題數的績效壓力，而持續進行追蹤覆核審計並再次提出審核意見；而追蹤覆核審計雖然帶給受查機關不小的壓力，但能否發揮預期效果，仍須視查核議題的改善可行性，一旦審計人員提出的缺失內容，需要耗費很多時間及經費，才能完成改善，無論查核頻率再高，可能也難以發揮追蹤覆核審計的效果。

關鍵字

績效審計；公共課責；審核意見；追蹤覆核審計；受查機關；卡方檢定

並列摘要

In response to the development of New Public Management, performance audit, which serves as a major tool for government performance management, theoretically should induce auditees to practice improvements on their successive administrative polices or operation procedures through audit findings, thus enhance the overall public accountability of the government. However, according to years of foreign experiences on practicing performance audit, causality did not completely exist between the establishment of performance audit system and its effectiveness. Some researches even discovered that the efficacy of performance audit failed to meet its expectation. It has been 13 years since the aggressive implementation of performance audit by the National Audit Office in 2009. This research attempted to discuss the relationship between the performance audit and government policy efficacy of Taiwan and the various factors that may influence whether performance audit could demonstrate its efficacy or not. Previous empirical researches regarding performance audit discussed this issue from the 4 perspectives of “enhancing financial efficiency”, “the level of perception from auditees on the effectiveness of performance audit”, “the responses and improvements from auditees regarding audit findings” and “public debate or media coverage due to performance audit”. These researches have proved that when examining the efficacy of performance audit, 7 factors including “quality of audit findings”, “deficiency types in audit findings”, “level of usage by external users on audit findings”, “auditee’s attitude towards performance audit”, “audit frequency”, “level of hierarchy of the auditee” and “consistency on audit findings and the policy direction of the auditee” would indeed have influence on the effectiveness of performance audit. This research practiced archive researches on the audit findings and auditee responses regarding the Council of Indigenous Peoples, the Environmental Protection Agency and the Ocean Affairs Council from the National Audit Office in 2014 to 2019. This research discovered that 90% of the audit findings were agreed by auditees and most auditee personnel agreed to the audit findings, indicating that the performance of performance audit of Taiwan has achieved a certain level of effectiveness. However, the percentage of auditees practicing improvement measures decreased to 67%, indicating that although auditee personnel conceived performance audit as effective, it was not necessary for audtiees to alter their policy implementations according to audit findings. On the other hand, according to the results of Chi-square tests, different types of audit and deficiency types in audit findings indeed possessed evident influence on whether auditees practiced improvement measures or not. Audit findings that bear the features of compliance audit and performance audit enabled auditees to propose possible concrete improvement measures. As for deficiency types in audit findings, audit findings that were related to “insufficient budget or personnel” and “incomprehensive supervisory and assessment mechanism” were conducive to auditees in fighting for more relevant personnel budgets or financial subsidies, and therefore had higher possibilities of proposing improvement measures. On the contrary, audit findings that were related to “incomprehensive legal regulations” had lower possibilities of proposing improvement measures by auditees. Finally, according to interviews, audit and auditee personnel both believed that follow-up audit indeed casted significant pressure on auditees and are conducive on inducing them to hasten their improvement. However, contents from interviews failed to be reflected on actual auditee responses. This research discovered that follow-up audits tended to stress on the importance of audit agendas and the level of outside attentions instead of “room needed to be improved”. Even if auditees have proposed concrete improvement measures in their previous year audit responses, audit personnel might continue to practice follow-up audits and repeatedly propose audit findings due to external pressures or the performance pressure on number of disclosure agendas in Annual Final Statements Reports. Although follow-up audits would cast significant pressure on auditees, its performance depends on the improvement possibility of the audit agenda. If the deficiency pointed out by audit personnel requires significant time and budget to improve, it is difficult for the effectiveness of follow-up audit to be demonstrated no matter how frequent audits were being practiced.