本文應用修改 CS 及 DGV 的實證模型,檢視貸款成數與違約機率之間是否呈現正向相關,來解釋房屋貸款市場資訊不對稱的現象。研究結果發現,由 CS 模型迴歸分析得知,不論是否控制利率,皆顯示房貸市場借貸雙方存在著資訊不對稱之情況,且貸款年限愈長與貸款金額愈大,其違約機率便愈高,顯示此時道德風險較大。另由 DGV 模型迴歸分析得知,在低貸款金額下,貸款成數隨貸款金額增加而減少,表示此時貸款人的逆選擇問題較小;擔保品價值較低時會提升貸款成數,此時貸款人的逆選擇問題較為嚴重。為了探討貸款金額大小與利率高低的搭配對於資訊不對稱的影響,我們將樣本區分為四種組合進行實證。結果顯示,銀行與貸款人之間除低貸款總額高利率組別外,其餘組別皆存在著資訊不對稱的問題。貸款人性別對高貸款成數的影響並不顯著,中、南部地區貸款人傾向於選擇貸款成數較高的契約,而貸款利率確實為借貸雙方中資訊不對稱問題的主要因素之一。
In this paper, we modify CS and DGV empirical model to explore the problems of asymmetric information between banks and borrowers in Taiwan''s house mortgage markets. The results show that by the CS model regression analysis, whether control interest rates, mortgage market is shown the existence of information asymmetry between bank and borrowers. By the DGV model regression analysis, the borrower''s adverse selection problem is smaller in the low loan amount; the borrower''s adverse selection problem is more serious in low loan-to-value. In order to explore how the size of the loan amount combine with interest rate whether effect information asymmetry. The result shows that interest rates is a major factor in the problem of information asymmetry.