基於(1)國內金融機構逾期放款金額嚴重，傳統不良債權(NPL)清理模式已不敷「當今社會脈動」使用，(2)政府動用納稅人龐大稅金投入金融重建(RTC)，(3)新知識經濟浪潮已轉變國人對傳統財富認知觀念，及(4)創新思維之考量，本研究以台中市營建產業個案為對象，從交易成本、代理成本及資源基礎觀點深入探討金融機構將不良債權直接讓售予資產管理公司(AMC,準外化模式)與將不良資產(NPA)採專業代工途徑(OEM/ODM,準內化模式)，兩項清理模式其最終執行成效之比較。長久以來，金融機構逾期放款中，以不動產為擔保品者居大宗，若能論證而得創新清理模式，並擇適推行，消極面可避免龐大不動產商品透過各種拍賣方式進入市場，擾亂正常交易次序與價格，削減了政府原欲挽救金融產業與刺激低迷不動產市場所頒佈之多項律法、財稅與行政措施之功效；同時也為營建業另闢一條創新經營之道。積極面將有助於提升金融機構債權收回速度與成數，減少營業損失，間接降低政府投入金融重建基金之規模，減輕全民負擔。本研究全程以創新思維出發，以個案實例為題材，經採行個案研究法並引證交易成本、代理成本及資源基礎理論與觀點，作為個案研究論證基礎，嘗試另覓傳統思維以外的創新見解。本研究經過個案分析，推導命題如下:在其他情況不變下，(1)不確定因素(如環境面與技術面)，(2)資訊不對稱因素，(3)代理問題所導致之監督成本因素，及(4)處理不良資產之核心能耐，會影響金融機構不良債權與不良資產清理模式之選擇；而採取不同的不良債權與不良資產清理模式會影響清理成效。研究論證顯示，金融機構不良債權中之不動產擔保品，若採「準外化」清理模式，對原始直接債權銀行言，其最終成效(債權收回速度與收回成數)將傾向劣於傳統「完全外化」清理模式。若採「準內化」清理模式，其最終成效則傾向優於「準外化」、「完全外化」清理模式。 Over the past five years, especially peak in 2001, delinquent ratio of the financial institutions in Taiwan has received a historical-record amount of attention from a broad range of the country. Evidence of this attention takes many forms, the most visible being the implementation of RTC, set-up by the government recently with amount of NT$140 billions initially, to recover the poor financial assets. The second way out to minimize the delinquent ratio is permission to incorporate the 「Asset Management Company, AMC」, which has privilege to acquire the Non-Performing-Loan (NPL) of the financial institutions officially. And in the period of 2002~2004, total amount of NT$ 477.8 billions’ NPL were transferred to various AMC.From the perspectives of transaction cost, agency cost, and resource-based views, the study through a comprehensive case will stress on the correlational performance analysis between the models of AMC (so-called quasi-externalization) and OEM/ODM (so-called quasi-internalization) in dealing with the resolution of NPL.As a result, the study has reached the following propositions:Holding all else equal, the ? uncertainties of the economic environment and technological employment ? information asymmetries ? agency problems ? core competence, will influence the decision-making upon selecting the NPL resolution models. And the final performance of quasi-internalization model has a higher opportunity to better than that of quasi-externalization model.
