在IFRS 9下信用風險顯著增加之評估-以債券為例

台灣於2018年1月1日正式啟用國際財務報導準則第9號「金融工具」，減損模式由「預期損失模式」取代過去IAS 39「已發生減損模式」。預期損失模式要求企業評估金融工具自原始認列後「信用風險是否有顯著增加」且「是否屬於低度信用風險」，以兩準則判斷應認列12個月亦或是存續期間之預期信用損失提列備抵損失。有鑑於台灣尚無信用風險顯著增加相關的衡量以及判斷依據相關文獻，並考量資料取得之問題，選擇國外信用違約交換（CDS）作為信用風險顯著增加之釋例做探討。本文將信用風險定義成違約機率，致力於提供量化判斷指標，並簡化既有違約機率模型輔以彭博內建違約機率模型驗算其可行性後，運用市場交易資訊如CDS指數和外部評等報告輔助判斷。最後回歸台灣債券市場，利用台灣證券櫃檯買賣中心提供公債及公司債殖利率做適度轉換後可運用於做判斷依據之例子。

關鍵字

國際財務報導準則第9號；違約機率；預期損失；信用違約交換；信用風險

並列摘要

IFRS 9 “Financial Instruments” will be officially effective from 1 January 2018, and the impairment model of IAS 39 will be replaced by the “expected loss model”. The expected loss model requires companies to evaluate the credit risk of an instrument, and to judge “whether the credit risk of has increased significantly” and "whether it is a in a low credit risk profile" after the original recognition. These two criteria mentioned above are used to determine whether 12 months or lifetime amount of loss allowance to be recognized. In view of the fact that there is very few literature in Taiwan to discuss how to capture or evaluate significant deterioration, and considering the accessibility of data, in this paper choose credit default swap (CDS) as an instance discussing the issue. In this paper, credit risk is defined as the probability of default, aiming at providing quantitative judgment guideline. Compare the default probability (PD) deriving from the simplified model revised by this paper to Bloomberg's model to measure the feasibility of the previous one. Then, utilize the market data, such as CDS index, external rating report as criteria to compare with the PD of an instrument. Finally, back to the Taiwan bond market, the paper will show an example of what informant can be use or how to use to evaluate PD of bonds by using public information provided by Taipei Exchange.