台灣的銀行家數過多，借款企業的額度動用率又低於七成。因此，欲探討企業與銀行建立的往來密切程度，需要納入多重、而非單純的往來關係之影響。至於此多重的往來關係，則須兼顧第一(所有權類型)與第二種(往來家數)衡量方法。具體言之，本實證模型擬將借款企業與各所有權類型銀行(主要分成公營、民營金控、民營非金控及外資銀行四種類型)的往來程度(以借款企業向公營、民營金控、民營非金控及外資銀行的個別融資金額對該企業總資產的比率來衡量)納入，係因台灣十多年來的金融環境與金融變革，導致商業銀行的經營態勢發生重大且明顯的變化，需要適度掌握其對借款企業投融資決策的影響；另一方面，實證模型還將多重往來關係(以往來家數衡量)對往來密切程度的影響列入，觀察其能否提升對借款利差與資本投資增幅的解釋；也就是建立「往來家數對各所有權銀行借貸密切程度的迴歸」。基於完整起見，謹臚列本研究之主要目的如次： 1. 首先，利用t檢定和無母數的Wilcoxon rank sum方法，驗證不同信用風險和借貸密切程度差異之所有權類型銀行，其借款企業在往來銀行的家數、借款利差、資本投資增幅、營運現金流量、信用風險、信用評等狀態、規模、負債比率、獲利比率和組織型態的平均數和中位數是否存在顯著的統計差異。 2. 接著，著眼於多重往來關係對企業借款利差和資本投資增幅的影響？除基本模型外，第二項目的旨在利用借款企業的信用風險和營運現金流量，與主要構面分別建立的交叉項，檢視其對借款利差和資本投資增幅的影響，並藉由各變數的正負向關係，推測債權銀行的授信行為會否隨著多重往來的程度而呈現顯著差異？ 3. 台灣特殊的金融環境，使得企業的「往來銀行家數」可能無法單獨描述多重往來關係，而要藉由各所有權銀行的融資金額對該企業總資產的比率(以下簡稱融資比率)，才能有效掌握借款企業依賴多重往來關係的程度。具體言之，第三項目的係透過「往來家數對實證模型中各所有權銀行融資比率之迴歸係數」的架構關係，進一步檢視其能否提升對借款利差與資本投資增幅的影響。在穩定性測試(robust test)方面，實證模型除利用三個虛擬變數描述四種所有權類型外，還藉由個別申貸案件的債權銀行是最大、主力(第二至第四大債權銀行)與非主力(第五至第二十大債權銀行)(李思慧，2010)(以兩個虛擬變數衡量)，來反映借貸密切程度；也就是建立「往來家數對此二虛擬變數在實證模型之迴歸係數」的架構關係，進一步檢視其能否提升對借款利差與資本投資增幅的影響。 This study examines the effect of borrower’s credit risk rating on banks’ credit behavior after the financial restructuring in Taiwan. The evidence shows that risky firms experience higher loan rate. And not only bank with “the write off large bad debt” but also banks with “large loss from credit card crisis” significantly price higher loan rate for their customers. However, bank with “large loss from credit card crisis” still lend lower rate for the high credit risk borrowers; Because of deregulation and the involvement of foreign banks, Taiwan’s financial industry become more and more competition. In this situation private domestic banks tend to provide lower rate for their customers regardless credit risk. On the contrary, foreign banks price the rational loan rate depending on borrowers’ credit risk. Then borrowers with closer lending relationship pay low loan rate even for risky borrowers. The second purpose of this study is to investigate the effect of banks’ credit behavior on firms’ investment and financing decisions. The result shows that firms borrow from bank with “the write off large bad debt” or banks with “large loss from credit card crisis” have higher sensitivity of investment to cash flow. Firms with relationship with different bank ownership types also affect firms’ investment and financing decisions. Finally, the evidence suggests that firm’s credit risk rating and ratings upgrade or downgrade also influence the sensitivity of investment to cash flow. 研究期間：10008 ~ 10107
