經濟環境的快速變遷與激烈競爭，造成企業經營風險與不確定性大幅增加，使得銀行於企業放款業務方面所需承擔的風險亦大幅提升，銀行除追求不斷成長的業績與獲利外，更應掌控涉險部位的風險程度與預期損失。之前已有許多研究提出如何衡量企業的違約機率，但是銀行更關心的是違約企業的償還表現，若違約企業償還表現愈差，則銀行所需承擔的損失愈多，因此本研究提出一個預測違約企業償還率的模型，希望可以藉此讓銀行於企業違約前或違約當時即可估計該筆違約放款的預期損失。另一方面，本研究亦探討違約機率與總體經濟對償還率的影響，並於估計違約企業償還率時，將此兩因素同時納入考慮。 根據實證結果得到以下結論： 1.企業的違約機率與違約後償還率為反向關係，違約機率愈高償還率愈低；反之，違約機率愈低償還率愈高。 2.利用台股指數報酬率與經濟成長率為總體經濟的替代變數，藉以衡量總體經濟景氣與違約企業償還率之關係，結果發現景氣愈佳償還率愈高；景氣衰退則償還率愈低兩者為反向關係。 3.違約企業償還率預測模型中包含的顯著變數有公司規模、經營績效、違約機率、核貸金額、額度使用率、擔保品平均償還率、銀行往來家數與前期台股指數報酬率，除經營績效、核貸金額與銀行往來家數不具顯著解釋能力外，其餘變數皆具有顯著解釋能力。 4.影響違約企業償還率最重要的因素為擔保品，總體景氣僅為次要因素，因此銀行的授信政策仍是左右債權回收的主要因素，授信政策嚴謹、擔保品徵提充足，債權回收愈有保障。 The rapid changes and violent competitions of the economy have not only let the enterprise face more risk and uncertainties but also let the banks have to take more risk on the commercial loan. Therefore, besides looking for the outstanding accomplishments and profits banks need to control the degrees of exposed risk and expected loss. There have been a lot of researches about how to measure the probability of default. But there is another thing that the banks pay more attention to. That is the recoveries from the enterprise that has defaulted. If the recoveries are low, the losses of the bank will be a lot. We attempt to fount a model to estimate the recovery rate. We also hope that banks can use the model to measure the loss of default before an enterprise really defaults. Additionally, we also discuss the relationship between default rate and recovery rate and the relationship between economy and recovery rate. When we establish the recovery predict model, we will consider default rate and economic situation. According to the results of empirical study, the conclusions are as follow: 1. The relationship between the default rate and the recovery rate is negative. When the default rate is low, the recovery rate is high. In other words, when the default rate is high the recovery rate will be low. 2. Using the stock return and economic growth rate as the proxy variables of the economy. We find that when the economy blossom the recovery rate is high; when the economy decline the recovery rate is low. 3. The obvious variables in the model are the scale of corporate, default rate, use ratio, average recovery rate of collateral, numbers of correspondent banks, and stock index return. 4. The most important factor that influences the recovery rate is collateral. Economic situation is the second important factor. Therefore, the policy of making loans is the most important thing that determine the recovery rate. If banks request more collaterals banks will suffer less uncertainty.
