Įmonių reitingavimas : Lietuvos bankų patirtis

Direct Link:
Collection:
Mokslo publikacijos / Scientific publications
Document Type:
Straipsnis / Article
Language:
Lietuvių kalba / Lithuanian
Title:
Įmonių reitingavimas: Lietuvos bankų patirtis
Alternative Title:
Corporate rating: experience of Lithuanian banks
In the Journal:
Pinigų studijos. 2007, Nr. 1, p. 86-108
Keywords:
LT
Įmonės. Bendrovės / Companies. Enterprises.
Summary / Abstract:

LTStraipsnyje nagrinėjami teoriniai ir praktiniai įmonių reitingavimo aspektai, įskaitant Bazelio bankų priežiūros komiteto dokumentuose, Europos Parlamento ir Tarybos direktyvoje 2006/48/ED ir skolininkų reitingavimui skirtoje literatūroje nurodomus reikalavimus bankų taikomiems įmonių reitingavimo modeliams. Nagrinėjami vertinamų rodiklių atrankos, reitingavimo procedūros pasirinkimo ir taikymo, reitingų peržiūros, reitingavimo modelių patikimumo vertinimo pagrindiniai aspektai, reitingų taikymo sritys. Apžvelgus "geros praktikos" reikalavimus, vertinamas Lietuvos bankuose atliekamas įmonių reitingavimas ir pateikiami siūlymai dėl jo pagerinimo. [Iš leidinio]Reikšminiai žodžiai: įmonė; Reitingas; Reitingavimo modelis; Corporate; Rating; Rating models; Įmonė; Company; Rating model.

ENThe paper describes the theoretical and practical aspects of large corporate rating in banks. The first part of the paper discusses the whole rating process and the theoretical aspects of the best rating practices. The second part analyses the corporate rating practices in the Lithuanian banks. The first part of the paper starts with discussing general aspects of rating models; differences between borrower and exposure ratings are explained. After describing the differences between the two, the general corporate borrower rating process is described and depicted. The theoretical aspects of the rating process described in the first part of the paper can be divided into three sections, namely the building of a rating model, application of the rating model, and the use of the rating model. To build a rating model, one needs quantitative and qualitative data describing borrower credit risk. The paper describes the main indicators used in credit risk rating models, compares them with those of international rating agencies. After deciding on possible rating criteria, one needs to select the most important or the most predictive rating criteria and set a rating procedure, i.e. how an individual rating criterion shall be joined to the final rating. The paper depicts the steps in selecting a rating procedure and describes the main differences among rating models based on purely statistical methods and those purely on expert based methods. Next, choosing the number of rating grades is described. Different banks use a different number of ratings for defaulted and non-defaulted borrowers. The paper discusses the reasons for choosing a different number of rating grades and stresses that there is no one best solution in deciding on the number of rating grades.Further in the paper, point-it-time and through-the-cycle rating models and their differences are described. The paper discusses who usually assigns ratings in banks (credit analyst, client executives or both), also the pros and cons for choosing one or the other approach to assigning ratings. Before concluding the first chapter of the paper with rating model validation and possible uses of ratings in everyday business of banks, PD estimation for rating grades is described. The second part of the paper discusses how borrowers are rated in the Lithuanian banks and analyses whether this is done according to the best international practices requirements described in the first chapter of the paper. The analysis of borrower rating in the Lithuanian banks has led to the following conclusions and recommendations to the Lithuanian banks: 1. Banks in Lithuania apply the same rating models as their parent banks or the models developed on their own. 2. Banks using borrower rating models should not consider collateral as one of the rating criteria. 3. The number of risk grades should be sufficient to differentiate borrower risk. 4. All banks in Lithuania must validate their rating models. 5. Banks should define defaulted borrowers and set at least one rating grade for these borrowers. 6. Banks should clearly state first of all to themselves which type of rating model (point-in-time or through-the-cycle) they apply, as this has impact on PD estimation and further risk management in which ratings are used. 7. Banks applying exclusively an expert based rating procedure must implement necessary internal control tools to assure the consistency of their ratings.8. All rating models should include rating override possibilities as the standard rating procedure is not applicable to all borrowers or not in all cases. 9. Banks should collect and store information on all rating changes, as this information is necessary for further model development. 10. Banks must ensure that ratings are reviewed in time. 11. Ratings should be assigned by bank experts not involved in risk taking. 12. Banks should set the rating review frequency according to the risk of borrowers, i.e. a riskier borrower’s rating should be reviewed more frequently. 13. Banks should use rating data more widely in everyday business and risk management. 14. Banks should start thinking about rating not only borrowers but also exposures. 15. Banks should disclose to interested parties more information about their rating models, as required in the Basle II document. 16. Finally, not only those banks which apply internal ratings based capital estimation methods, but all the other banks as well should set their rating models and processes according to the best international standards. [From the publication]

ISSN:
1392-2637; 1648-8970
Related Publications:
Permalink:
https://www.lituanistika.lt/content/14586
Updated:
2018-12-17 12:00:37
Metrics:
Views: 27    Downloads: 2
Export: