| Pub. Date | : June, 2020 |
|---|---|
| Product Name | : The IUP Journal of Financial Risk Management |
| Product Type | : Article |
| Product Code | : IJFRM10620 |
| Author Name | : Brian Barnard |
| Availability | : YES |
| Subject/Domain | : Finance Management |
| Download Format | : PDF Format |
| No. of Pages | : 38 |
Rating migration variation or volatility, as rating migration uncertainty, is a real-life phenomenon that can be measured empirically. The study extends reduced form bond valuation models based on rating migration (matrices) by allowing variation in the rating migration matrix, as opposed to variation of the rating migration matrix. This is seen as a more accurate and practical modeling of downgrade risk and default risk, as including rating migration uncertainty. Rather than stating a number of possible rating migration matrices, commonly representing certain scenarios, each entry of the rating migration matrix is allowed to vary by a certain extent. The resultant price probability distribution is delineated by searching the minimum and maximum possible price through optimization. The study models bond value of individual issues as well as portfolios. The cost of downgrade or default is delineated by calculating possible future bond or portfolio prices at a certain point in time, and comparing the output price probability distributions with conventional bond valuation model prices. Because variation in the rating migration matrix is permitted, the model is better able to account for and discount the cost of downgrade or default. The outcome of the study suggests that rating migration uncertainty and variation, as a component of default risk and downgrade risk, may very well be a significant factor of bond value. Prices calculated in this way may deviate from conventional prices. Furthermore, portfolios do not necessarily diversify price uncertainty and variation due to rating migration uncertainty and variation.
Varying, as opposed to fixed or static, rating migration matrices over the investment horizon, are both practical and empirical. Some studies allude to them (Bangia et al., 2002; Hamilton and Cantor, 2004; Fei et al., 2012; Huang and Huang, 2012; and Barnard, 2019a), and the phenomenon can be empirically measured: Barnard (2019a) delineates integer migration and portfolio migration, and notes that it resembles (the simultaneous application of) multiple rating migration matrices. Rating migration matrix variation can be empirically measured as rating migration matrix drift: the variation in empirically measured and published rating migration matrices, from point in time, to point in time. And this would correspond to changes in outlooks.