灰色多準則決策於租賃業汽車貸款之風險評估與管理

Grey Multiple Criteria Decision Making for Risk Assessment of Auto Loans in Leasing Industry

金融機構為增加放款市佔率，紛紛提高額度、降低利率及鬆綁條件，卻也同時提高呆帳的風險。個案公司為台灣租賃業龍頭，車貸為其重要業務。貸款績效的良窳在於對逾放的管理，因此發展風險評估模式，降低風險對個案公司甚至對整個租賃業，都是重要的議題。本研究邀請專家確認影響風險評估之主要因素，也由於汽車貸款風險評估屬灰色系統問題，因此進一步以灰關聯分析為基礎發展評估模式。對評估模式而言，特異度代表實際為呆帳資料中，被判斷為呆帳的比例，因此特異度愈高的預測模式對個案公司而言更具實用性。所發展模型與其他資料探勘技術相較下，的確具有較佳的整體正確率與特異度，因此顯示出所發展模型於風險評估之優越性，亦完全符合租賃業的實務需求。此一模式可有效輔助個案公司於車貸之審查，並可簡化現有審查流程，降低呆帳風險，達成永續經營的目標。

 

This paper investigates the substitution effect between corporate tax avoidance and debt, and examines its sensibility of three firm attributes including size, profitability and credit constraint, using a sample of Taiwan listed corporations in 2005-2015. We find that tax avoidance is significantly negatively associated with leverage measured by total debt, which is partially consistent with Graham and Tucker (2006), Rao and Yu (2013) as well as Huang and Wang (2015); while, tax avoidance is significantly positively associated with leverage measured by long-term debt, which is different from Rao and Yu (2013) and is consistent with Bradley et al. (1984) as well as Downs (1993). Furthermore, the relation between tax avoidance and leverage is significant and negative for large firms when leverage measured by total debt, and is significant and positive for smaller firms when leverage measured by long-term debt. In addition, the relation between tax avoidance and leverage is significant and negative for less profitable firms when leverage measured by total debt, and is significant and positive for the profitable firms when leverage measured by long-term debt. Lastly, the relation between tax avoidance and leverage is significant and negative for firms with less favorable prior credit ratings when leverage measured by total debt, and is significant and positive for firms without credit ratings when leverage measured by long-term debt.