Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.492274
Title: An empirical analysis of the nature and incidence of credit risk in the Turkish banking sector : a balance sheet approach
Author: Ayaydin, Hande
ISNI:       0000 0001 3433 8040
Awarding Body: University of Manchester
Current Institution: University of Manchester
Date of Award: 2008
Availability of Full Text:
Access from EThOS:
Abstract:
This thesis examines credit risk in the Turkish banking sector. The specific focus is an in-depth analysis of the detenninants of changes in the future values of basic on-balance sheet assets, securities, claims on banks, credits, and a contra-asset account, past due loans, as well as off-balance sheet assets, namely guarantees and warranties, derivative financial instruments and repurchase transactions. The thesis jointly considers changes in the values of both on- and off-balance sheet assets which arise from the evolution of micro economic (bank-specific) and macroeconomic variables between the years 1990 and 2005. Although the Basel Committee highlights the importance of credit risk in all on- and off-balance sheet assets, the academic literature customarily focuses only on the credits and past due loans portfolios. Furthennore, existing academic literature concentrates almost exclusively on developed as opposed to emerging economies. We focus upon Turkey for a variety of reasons. In particular, we consider it to be a good example of an emerging economy which has undergone two recent major financial crises, in 1994 and 2001, which significantly impacted upon the banking sector. The analytical approach taken also differs from previous studies in that prior literature undertaking analysis of credits and past due loans concentrates almost exclusively on the effects of the macro economy on these portfolios, while the literature relating to off-balance sheet assets focuses much more on the effects of bank-specific variables. However, in this thesis on- and off-balance sheet assets are analysed in conjunction with each other in order to provide us with a comprehensive understanding of the credit risk in the sector. As a consequence, both bank-specific characteristics, and macro variables are employed in the estimates. The estimations are undertaken using both panel data and time series methodologies. The heterogeneity of the individual banks was taken into account through panel data analyses. Both static (fixed, random effects, Prais- Wins ten and Tobit models) and dynamic (one-step robust Arellano-Bond) estimations are used. Furthennore, the credit risk of the deposit money banks is examined with higher frequency data by means of time series analysis (employing autoregressive 12 distributed lag). In this manner, the interim effects of the evolution of both micro and macro variables on the assets of the banking sector are able to be considered. The analyses concludes that it is excess growth in the credits portfolio of the Turkish banking sector which decreases the quality of borrowers and increases the amount of past due loans and credit risk in the sector. Moreover, credits appear to be extended in order to control the amount of past due loans. Effectively, banks appear to try to postpone writing-off past due loans as losses. However this strategy, known as forbearance lending, does not prevent the inevitable and such loans tum into losses after two years. The change in gross domestic product is the most robust explanatory macroeconomic variable in the panel data analysis. As the economy grows banks invest more in securities, and extend more credits. The results are robust to bank size, different interest rate variables, type of banks and different time periods. Furthermore, when the results are controlled for the bank size we find that smaller banks are those extending more credits which tum into past due loans. Larger banks with higher levels of buffers of total assets prefer to invest in securities portfolios. This excess credit growth by small banks increases the vulnerability of the sector to macroeconomic shocks and increases its credit risk. The results also support the perspective that the Turkish banking sector is dominated by deposit money banks. The signs and significance of the variables do not change even when the investment and development banks are excluded from the data set. From a policy perspective, the results also highlight the necessity of the recent restructuring process which has been undertaken in the Turkish banking sector. This has served to decrease credit risk exposure in the sector. The restructuring process has affected mostly the securities and credit portfolios. Before this restructuring, the majority of the funds available for the banking sector were diverted to securities portfolios, and in addition there was excess credit growth in this time period. The time series analysis presents that after the restructuring process, bank managers pay more attention to the growth of past due loans when extending new credits. However forbearance lending is still a problem. Moreover, we find that the problems in the credits portfolios of banks taken into the savings deposit insurance fund and/or closed increase the credit risk in the sector. Therefore, the restructuring process was 13 inevitable and decrease part of the credit risk but there are yet some points to be improved. The results indicate that credit risk exposure of the Turkish banking sector is more than would be calculated simply by looking at the banks' on-balance sheet credit portfolios. When the economy is in recession or there is financial turmoil, banks are reluctant to lend via extending more on-balance sheet credits. Banks and the beneficiaries prefer to employ off-balance sheet credits, namely guarantees and warrantees in these circumstances. Moreover, smaller banks with less liquid assets and more on-balance sheet credits and past due loans, appear to extend more offbalance sheet credits and increase the risk structure of the sector. Furthermore, the analysis undertaken on repurchase transaction portfolios indicates that the funds which are diverted into securities portfolios between 1990-2000 period are employed in repo transactions, both for liability management purposes and also to meet the bank's liquidity demand. This makes the sector highly vulnerable to interest rate changes. Finally, we find that derivative financial instruments are employed to hedge against interest and exchange rate risk rather than credit risk.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.492274  DOI: Not available
Share: