• Size: 359.2 KB
  • Uploaded: 2018-10-11 14:28:32
  • Status: Successfully converted

Some snippets from your converted document:

M A T H E M A T I C A L E C O N O M I C S No. 7(14) 2011 VINTAGE ANALYSIS AS A BASIC TOOL FOR MONITORING CREDIT RISK Paweł Siarka Abstract. Among many tools used by bankers in the process of credit risk management, vintage analysis is the most often applied. Its simplicity and clarity of interpretation of the results means that banking professionals call it “basic analysis”. In this article, the concept of vintage analysis is presented, along with the right way to get interpretations of results. The author demonstrates the usefulness of vintage analysis in the context of the back- testing procedure recommended in New Basel Capital Accord. In addition, there is also a discussion of an aspect of the limit’s structure as a part of the credit risk management process. Keywords: credit risk, retail loans, probability of default. JEL Classification: C13, C16, D81, G21, G32, G33. 1. Introduction Bank employees took the term “vintage” directly from the world of wine. For a long time wine gourmets have been evaluating their noble beve- rages based on both the species of grapes and the year in which the grapes were harvested. Exactly the year in which grapes ripened has gigantic mea- ning for the depth of wine’s taste. It is known that poor insolation during a vegetative season in a given year can be a cause of the low sugar content, which in consequence has a gigantic influence on the final taste. Thus, for many years experts have been creating vintage tables, from which one can read a note determining the quality of a given wine from a particular year. Finally, based on a drawn up table, we will know whether a given wine should be stored longer in order to get the optimum taste, or if it should be drunk, or what is worse – if it should have been drunk much earlier. Wine lovers, who are quite common among bankers, have noticed some analogy between the variable quality of the wine from a given year and Agnieszka Stanimir Department of Econometrics, Wrocław University of Economics, Komandorska Street 118/120, 53-345 Wrocław, Poland. E-mail: agnieszka.stanimir@ue.wroc.pl 214 Paweł Siarka variable in time quality of credit portfolio built by the bank in a given year. It turns out that loan production performed in a given time can be success- fully described with the use of vintage tables. Bank vintage analysis is now a popular tool for managing credit risk. Due to its considerable functionality, it allows for the ongoing monitoring of risk level, as well as historical levels, providing a complex image of the phenomenon’s course in time. Therefore, it often constitutes a starting point for further analysis based on the analysis of time series, providing a broad spectrum of opportunities of analytical risk assessment. In the literature on the subject, the vintage analysis has been frequently presented as a universal source of knowledge on many areas related with the risk management process. Vintage analysis is recommended by the Experian company, which is one of the world biggest consulting firms specialising in credit risk management processes. Burns, Stanley (2001) in their paper emphasise the advantages of vintage analysis regarding risk management of retail loans. The usefulness of vintage analysis is presented also by Breeden (2004), who proposed its use in the stress testing analysis. Anderson (2007) dealt with the issue of applying vintage analysis in the process of assessing the accuracy of forecasts. The author con- centrated in his paper on analysing the discrepancies between the forecasts of a number of default loans and their actual realisation. In his paper, Zhang (2009) drew attention to the probable character of a vintage curve, whose change- ability is significantly influenced by macroeconomic factors. In addition, he proposed a formal model of risk development using the Gaussian process. It appears that vintage analysis may be applied in the process of esti- mating capital requirements. In their paper, Ash et al. (2007) considered various approaches to the issue of estimating capital requirements on ac- count of credit risk regarding credit card portfolio. At the same time, they drew attention to the problem of the appropriate segmentation of credit portfolio. In the mentioned paper, the authors were researching two kinds of approach to segmentation. In the first of them, one was analysing fixed segments determined based on vintage analysis, where division was made according to the time of portfolio origination. In the second case, segmenta- tion was conducted based on scoring points, which were determined every certain period. As a result of the obtained research studies, they noticed a significant correlation between the length of portfolio life and the pro- bability of default, which speaks for the application of vint

Recently converted files (publicly available):