近年來，隨著政府財政壓力增加及公共建設投資減少，我國積極推動促進民間參與公共建設（Public-Private Partnership， PPP）制度（簡稱促參模式），然而，由於促參模式在國內推動過程中，相關契約履約階段仍有提前終止之情形發生，顯示我國在制度設計與執行管理上仍存在盲點與挑戰。過往研究多著重於促參推動策略與成效評估，對於契約提前終止之成因探討相對不足，且缺乏對既有案例多重因果關係之系統性分析。因此，本研究旨在針對國內促參案件提前終止之因素進行彙整與分析。本研究以財政部112年度委託研究計畫所彙整之終止案例為基礎，以能夠取得資料的28件終止案例作為主要研究標的。本研究利用「模糊集質性比較分析法（fsQCA）」，探討包含民間因素、政府因素、履約問題、外部環境因素、主辦機關差異及是否為BOT案等六大前因條件之交互作用，釐清導致契約走向「非合意終止」或「合意終止」，以及「政府為主要肇因」與「民間為主要肇因」之因果路徑。研究結果發現：第一，促參案件的終止並非單一因素所致。當案件出現「履約問題」時，不論民間機構財務投入程度為何，通常皆會走向仲裁或訴訟等非合意終止途徑。第二，在制度設計上，具備高資產專屬性的「BOT 模式」隱含極高的政府被認定為主要肇因之風險；若政府於前置行政作業（如用地交付、環評）發生延宕，極易在爭議中被判定違約。第三，若要成功判定民間機構負擔完全違約責任，必要前提為「政府方行政程序無瑕疵」，任何細微的行政延宕皆可能成為廠商主張「混合過失」的依據。基於上述結論，本研究建議主辦機關在推動大型BOT案時，應設立明確的前置行政作業檢核斷點，並導入初期爭議預警機制；民間機構則應避免以消極手段構成實質違約，並在面臨外部衝擊時，善用合意退場機制。;In recent years， Taiwan has actively promoted the Public-Private Partnership (PPP) approach to attract private investment and enhance public services. However， the frequent early termination of PPP contracts reveals challenges in institutional design and contract management. Previous studies lack a systematic analysis of the complex， multiple causes behind these terminations. This study aims to analyze the critical factors contributing to the early termination of PPP projects in Taiwan. Based on 28 termination cases， this study adopts the Fuzzy-set Qualitative Comparative Analysis (fsQCA) approach to explore the interactions among six antecedent conditions (private factors， government factors， performance problems， external environment factors， differences by authority-in-charge， and BOT models vs. non- BOT models)， and to identify the causal pathways leading to consensual and non-consensual termination， and termination liability. The empirical results reveal three major findings. First， the termination of a PPP project is rarely attributable to a single factor. Once contract performance issues arise， the project typically proceeds toward non-consensual termination through arbitration or litigation， regardless of the extent of the private partner′s financial investment. Second， from an institutional design perspective， the asset-specific nature of the BOT model entails a substantial risk that the government will be identified as the primary party responsible for project failure. If the government delays critical preliminary administrative procedures， such as land acquisition or environmental impact assessment approval， it is highly likely to be found in breach of contract during subsequent dispute resolution. Third， a prerequisite for assigning full liability for breach of contract to the private partner is that the government′s administrative procedures are entirely free from procedural deficiencies. Even minor administrative delays may provide sufficient grounds for the private partner to argue contributory or shared fault. Based on these findings， this study recommends that the competent authority should establish clear checkpoints for preliminary administrative procedures when implementing large-scale BOT projects and introduce an early dispute warning mechanism. Meanwhile， private partners should avoid passive conduct that may constitute a material breach of contract and should make appropriate use of consensual termination mechanisms when confronted with significant external disruptions.
