




































 

 

 
79 

© 2020 Conscientia Beam. All Rights Reserved. 

FINANCIAL RISKS IN TURKISH BANKING INDUSTRY: A PANEL DATA ANALAYSIS 
ON ISTANBUL STOCK EXCHANGE   

 

 

Cem Berk1+ 
Eyyup Arslan2 

 

1Kirklareli University, Turkey. 

 
2Independent Researcher, Turkey. 

 
  

(+ Corresponding author) 

 ABSTRACT 
 
Article History 
Received: 21 September 2020 
Revised: 5 October 2020 
Accepted: 16 October 2020 
Published: 28 October 2020 
 

Keywords 
Banking industry 
Exchange rate risk 
Financial risks 
Interest rate risk 
Liquidity risk 
Panel data. 
 
 

JEL Classification: 
C23; F31; G21; G32. 

 
Global price movements have been affecting markets dramatically in recent years. The 
changes in exchange rates, interest rates, and liquidity directly affect market value of 
firms. These risks are called financial risks and typically affect financial institutions. 
Many methods are developed to compute these risks. This study has a panel data 
analysis on 7 banks listed on Istanbul Stock Exchange. The motivation of this study is 
to investigate the relationship between financial risks (interest rate risk, exchange rate 
risk and liquidity risk) and market value of these banks. Many tests are available in the 
research such as VIF, AR Roots, Lag Length Selection Criteria, Cross Section 
Dependence Test, Delta Test, Unit Root Tests, Model Selection Tests, 
Heteroscedasticity and Autocorrelation Tests. Based on the tests, two way fixed effects 
model is developed. The results reveal that financial risks explain 29% of all price 
movements of commercial banks. The model is statistically significant. There is a 
positive relationship between liquidity and market value and negative relationships 
between interest rate risk and market value, and exchange rate risk and market value. 
The results are also consistent with the literature. The research is unique for the 
Turkish Banking industry and therefore is important academically as well as for risk 
management practice. Results show that banks operating in Turkey don’t properly 
manage financial risks. Macroeconomic dynamics and maturity mismatch problems in 
Turkey require great attention on financial risks. It is recommended that banks should 
operate with more risk management instruments such as financial derivatives and 
corporate risk management.  
 
 

Contribution/Originality: This study is one of the very few studies which have investigated the relationship 

between financial risks and market value of Turkish commercial banks. Most studies on financial risks have 

analyzed non-financial firms. And the studies on financial risks of financial firms primarily focus on profitability.  

 

1. INTRODUCTION 

Financial risk can be defined as the loss potential which may affect a company negatively to reach its goals and 

possibly results in a profit less than expected (Koroglu, 2019). This risk will lead to an increase in liabilities or a 

decrease in assets of a company. In the event of fluctuation of prices in the market, especially commercial banks are 

very fragile. Many techniques have been developed to measure this risk (Kocak, 2012).  

Price fluctuations in the markets have increased in the recent years. The fluctuations in exchange rates, interest 

rates, stocks and commodities directly affect firms. Because of these factors, firms are exposed to financial risks 

which are difficult to be managed (Yucel, Mandacı, & Kurt, 2007).  

Financial Risk and Management Reviews 
2020 Vol. 6, No. 1, pp. 79-87. 
ISSN(e): 2411-6408 
ISSN(p): 2412-3404 
DOI: 10.18488/journal.89.2020.61.79.87 
© 2020 Conscientia Beam. All Rights Reserved. 

 
 
 

 
 
 

 

 
 
 
 

https://www.doi.org/10.18488/journal.89.2020.61.79.87


Financial Risk and Management Reviews, 2020, 6(1): 79-87 

 

 
80 

© 2020 Conscientia Beam. All Rights Reserved. 

Companies work in the field of risk management to still obtain profits while protecting themselves from 

financial risks. Financial risks may lead to low financial leverage, negative changes in exchange rates, 

overdependence to a supplier, loss of primary customer, and loss of external investors (Chapman, 2006). 

Due to increased communication and information technologies, firms and individuals interact more closely. The 

advances in Internet led to developments in domestic trade. On the other hand, economic and political operations 

between countries and continents have increased. A negative situation in one part of the World may lead to several 

consequences in other parts of the World. These financial shocks stem from economic and non-economic reasons 

(Yanartaş, 2010).  

Examples of economic factors that is related to financial risk are inflation, interest rate, exchange rate, and 

market volatility. Non-economic financial risks may occur because of political and social events. The increase in 

financial instruments and information technologies has led to an increase in the number of financial risks that can 

occur. Firms develop new financial instruments to attract global capital, however increased leverage also come with 

greater financial risks (Uzak, 2019). 

The motivation of this study is to investigate the relationship between market value and financial risks of 

commercial banks. For this purpose, a panel data analysis of commercial banks listed on Istanbul Stock Exchange is 

done. The rest of this study is organized as follows. In the next chapter, several previous works in this field are 

given. In the main focus of the study section, the motivation and information on data set is available. The 

methodology section has information on the computation of variables and panel data regression. The results of the 

analysis and a brief discussion are available in solutions and recommendations part. Final remarks are available in 

the conclusion section of this study.  

 

2. BACKGROUND  

Most works on the literature of financial risks focus on non-financial firms. Most of the studies that focus on 

commercial banks have analysis on profitability.  

Unal and Altin (2010) analyzed the relationship between firm value and net foreign exchange position in the 

Turkish Automotive Industry. Reis, Kilic, and Bugan (2016) showed that GDP, leverage ratio, loans/deposits ratio, 

and market capitalization affect probability. The study is a panel data analysis. The research period is between 2009 

and 2013.Saldanli and Aydin (2016) analyzed 23 commercial banks. The research period is between 2004 and 2014. 

They found that shareholders’ equity / total assets, liquid assets / current liabilities, non-interest income / total 

assets, and interest income/ interest expense ratios all affect profitability of banks.  

Kok, Ekinci, and Ay (2017) studied the effect of financial risks with an ARDL model. The research period is 

1993-2015. They found a relationship between financial risks and non-financial firms. Senol and Karaca (2017) 

studied the effect of financial risks on non-financial firms. The analysis includes 35 firms. The research period is 

2008-2015. They found that there are statistically significant relationships between exchange rate risk and liquidity 

risk and Tobin’s Q and leverage and credit risk and market value. Topaloglu (2018) used 3 different models to test 

the relationship between financial risk and market value of non-financial firms.  Accordingly, there are statistically 

significant relationships between capital and liquidity risk and Tobin’s Q, credit risk and market value, and 

exchange rate risk and credit risk and price / earnings ratio.  

Senol, Oncül, and Buyer (2019) found that liquidity risk affect profitability positively, whereas credit and capital 

risks negatively affect profitability of commercial banks. The study consists of 19 commercial banks.  

 

3. MAIN FOCUS OF THE STUDY 

In this study, the relationship between market value of commercial banks and financial risks in these banks are 

analyzed. Financial risks are studied in three broad categories which are liquidity risk, exchange rate risk, and 

interest rate risk.  



Financial Risk and Management Reviews, 2020, 6(1): 79-87 

 

 
81 

© 2020 Conscientia Beam. All Rights Reserved. 

The research period is between 2010 and 2019. The reason of starting the research with the year 2010 is 

extremely high volatility in banking industry in the years 2008 and 2009 due to global financial crisis. The data 

used in this research is obtained from annual reports of the banks, Banking Regulation and Supervisory Agency, 

and Finnet database.  

The commercial banks analyzed in this study is given in Table 1. Albaraka Turk is excluded from the study 

because it doesn’t accept deposits. TSKB (industrial development bank of Turkey) is not included because it doesn’t 

have regular financial operations. ICBC Turkey and Q&B Finansbank are also not studied due to the the high 

volatility of these stocks related with Mergers & Aquisitions activities. 

 
Table-1. Banks analyzed in the research. 

No. Bank Ticker 

1 Akbank AKBNK 
2 Türkiye Garanti Bankası GARAN 
3 Türkiye Halk Bankası HALKB 
4 Türkiye İş Bankası ISCTR 

5 Şekerbank SKBNK 

6 Vakıfbank VAKBN 
7 Yapı ve Kredi Bankası YKBNK 

 

 

These banks are listed in Istanbul Stock Exchange. They have to make disclosures on financial risks such as 

exchange rate risk, interest rate risk and liquidity risk in their quarterly financial reports. The data on financial 

risks are obtained from these audited financial reports.  

The motivation of the study is to see whether there is a relationship between financial risks and market value of 

the commercial banks operating in Turkey. If there is such a relationship, a further analysis is what kind of financial 

risks affect the market price. This is important for academic purposes and for the banks in order to effectively 

manage these risks.   

 

4. METHODOLOGY 

The formulas to obtain variables used in the study to represent independent variables of financial risk; 

exchange rate risk, interest rate risk and liquidity risk and the dependent variable of market value are given in 

Table 2. These values are computed for each of the bank analyzed in this study.  

 
Table-2. Formulas to obtain variables. 

 Variable Abbreviation Formula 

Independent 
Variables 

Exchange Rate Risk EXC Net Foreign Exchange Position / 
Shareholders’ Equity 

Interest Rate Risk 
 

INT 
(Interest Rate Risk t – Interest Rate 
Risk t-1) / Interest Rate Risk t-1 

Liquidity Position LIQ Cash and Liquid Assets / Total Assets 

Dependent Variable Market Value MBV Market Value / Book Value 
 

 

Commercial banks disclose net foreign exchange position in financial statements. This is used in exchange rate 

calculations of this study. To account for the size effect, net foreign exchange position is divided to shareholders’ 

equity. Interest rate risk consists of the maturity mismatch in the balance sheet which is related with asset liability 

management and revaluation of off balance sheet positions. In the study, interest rate risk disclosed in the annual 

reports of the banks is used. For a meaningful calculation, the change in interest rate risk is used instead of the level 

of interest rate risk.  



Financial Risk and Management Reviews, 2020, 6(1): 79-87 

 

 
82 

© 2020 Conscientia Beam. All Rights Reserved. 

As liquidity variable liquidity position is used. Due to unique balance sheet structure of the banks, liquidity 

position is a meaningful indicator of liquidity risk. The banks with higher liquidity position are exposed to less 

liquidity risk.  

Panel data analysis is used study. The model allows to test the relationship between liquidity risk and market 

value, interest rate risk and market value and exchange rate risk and market value. The panel data regression model 

is given below. In the formula i stands for each commercial bank used in the study t stands for the time period, β is 

coefficient of independent variable, and Ԑ is the error term. 

MBVit =  αit+ β2it EXCit + β3it INTit + β4it LIQit + Ԑit + λt 

 

5. SOLUTIONS AND RECOMMENDATIONS  

First the data is tested for multicollinearity. The method used for this is VIF. The results are presented in 

Table 3. The maximum VIF value is 1.016518 which is less than five. Therefore, there is no multicollinearity 

problem for the variables used in this research.  

 
Table-3. Results for VIF. 

MBV 

Coefficient Variance Coefficient Non Central  VIF Central VID 

LIQ 4.878232 28.339460 1.016518 
EXC 0.000174 1.014890 1.012592 
INT 0.000180 1.133516 1.005991 

C 0.078783 28.132030 NA 
 

 

To test the validity of the model, AR Roots are analyzed. The results are given in Figure 1. Accordingly, all of 

the roots are inside the unit circle. This shows the model is appropriate for the analysis.  

 

 
Figure-1. AR Roots. 

 

Appropriate lag length has to be determined in panel data analysis. Lag length selection criteria are used for 

this purpose. The results are given in Table 4. Accordingly, 2 lengths are suggested by all (AIC, SC, HQ, LR and 

FPE) criteria.  
 
 
 
 
 
 
 
 



Financial Risk and Management Reviews, 2020, 6(1): 79-87 

 

 
83 

© 2020 Conscientia Beam. All Rights Reserved. 

Table-4. Lag length selection criteria. 

Lag LR FPE AIC SC HQ 

0 NA 0.017842 7.3253 7.479734 7.383892 
1 73.18276 0.006516 6.315117 7.087288 6.608078 
2 51.75661* 0.003481* 5.674263* 7.064171* 6.201592* 
3 21.54703 0.003808 5.728795 7.736441 6.490493 

 

 

Cross section dependence is also analyzed in this study. Since time period (10) is more than number of banks (7) 

Peseran CD is applied. The results are given in Table 5. Accordingly, for the variables MBV and INT there is no 

cross section dependence as the probability is greater than 0.05, whereas for the variables EXC and LIC there is 

cross section dependence.  

 
Table-5. Cross section dependence test. 

Variable Test Statistics Probability Value 

MBV -0.983 0.163 
EXC -1.899 0.029 
INT -0.805 0.211 
LIQ -1.672 0.047 

 

 

In panel data analysis, series are assumed to be homogenous. This assumption is tested with Pearson and 

Yagamata Delta Test. The results are shown in Table 6. As for all of the variables probability values are larger than 

critical value, slope coefficients are homogenous.  

 
Table-6. Delta test results. 

MBV EXC 

 

0.877 
 

1.049 

 

-1.446 
 

-1.728 

Prob 0.190 Prob 0.147 Olasılık 0.926 Prob 0.958 

INT LIQ 

 

-1.543 
 

-1.844 

 

0.777 
 

0.929 

Prob 0.939 Prob 0.967 Olasılık 0.218 Prob 0.176 
 

 

The variables used in the panel data analysis has to be stationary. Based on the results of cross section and 

homogeneity tests, Levin, Lin and Chu (LLC) test is used for MBV and INT while Bai and Ng PANIC test is used 

for EXC and LIQ. The results for LLC test is given in Table 7. According to LLC Test, the variables MBV and 

INT are stationary in level. The results of PANIC test are available in Table 8. Accordingly, the variables EXC and 

LIQ are not stationary in level but stationary in first difference.  

 
Table-7.  LLC panel unit root test results. 

Variable Statistics Probability 

Constant+ 
Trend 

Variable Statistics Probability 

MBV -2.42688 0.0076*** MBV -10.9845 0.0000*** 

INT -7.04066 0.0000*** INT -6.62184 0.0000*** 
 

 

 



Financial Risk and Management Reviews, 2020, 6(1): 79-87 

 

 
84 

© 2020 Conscientia Beam. All Rights Reserved. 

Table-8. PANIC panel unit root test results. 

 Constant Constant +Trend 

Level Statistics Probabilty Statistics Probabilty 

EXC     
 

-0.6888 0.7545 0.1877 0.4256 

 

10.3550 0.7358 14.9930 0.3786 

LIQ 
 

-0.8612 0.8054 -0.9543 0.8300 

 

9.4431 0.8017 8.9503 0.8342 

First Difference 

EXC     
 

1.8899 0.0294** 2.4545 0.0071*** 

 

24.0003 0.0458** 26.9882 0.0193** 

LIQ     
 

3.4017 0.0003*** 5.3716 0.0000*** 

 

32.0003 0.0040*** 42.4236 0.0001*** 

 
Table-9. Model selection results. 

 Test          Statistics p-value Hypothesis 

F-group_fixed 14.64063 0.000000 H0:Cross section Effect 

F-time_ fixed 19.84238 0.000000 H0: Time Effect  

F-two way_ fixed 17.88136 0.000000 H0:No Effect  

LM-group_random 12.52044 0.000403 H0: Cross section Effect 

LM-time_ random 35.68262 2.32E-09 H0: Time Effect 

LM- two way_ random 48.20306 3.41E-11 H0: No Effect 

Honda-group_ random 3.538424 0.000201 H0: Cross section Effect 

Honda-time_ random 5.973493 1.16E-09 H0: Time Effect  

Honda-twoway_ random 6.725941 8.72E-12 H0: No Effect 
 

 

In the analysis, an evaluation is made to choose between fixed effect and random effect models. The tests used 

for this purpose are F test, Breuch-Pagan LM test and Honda test. The results are shown in Table 9. According to 

F test results, group and time two way fixed effects model is suggested. However LM and Honda Tests suggest two 



Financial Risk and Management Reviews, 2020, 6(1): 79-87 

 

 
85 

© 2020 Conscientia Beam. All Rights Reserved. 

way random effects model. Considering data set of the study, two way random effects model is chosen for the 

analysis.  

Finally, error term is tested for heteroscedasticity and autocorrelation. Heteroscedasticity is tested with 

Breusch-Pagan-Godfrey LM Test. Autocorrelation is tested with Baltagi and Li, Born and Bretuing and Durbin-

Watson tests. The results are presented in Table 10. According to the results, there are both heteroscedasticity and 

autocorrelation problems in the model.  

 
Table-10. Heteroscedasticity and autocorrelation tests results. 

Heteroscedasticity 

Breusch-Pagan-Godfrey LMh_fixed 36.49401 0.000000 

H0: No Heteroscedasticity  
H1: Heteroscedasticity 

Autocorrelation 

Baltagi and Li (1991) LMp-stat 13.44682 0.000245 
H0: No Autocorrelation 
H1: Autocorrelation 

Born and Breitung (2016) LMp*-stat 22.13317 0.000003 

H0: No Autocorrelation  
H1: Autocorrelation  

Durbin-Watson 
Bhargava, Franzini and Narendranathan  

0.683223 

H0: No Autocorrelation 
H1: Autocorrelation 
 

 
Table-11. Model Results. 

Dependent Variable Method Data 

MBV 
Least Square Method 

White Period standard errors & covariance (d.f. corrected) 
2010-2019 

Independent Variable Coefficient Std Error t-Stat Prob 

LIQ 5.790447 2.852820 2.029727 0.0474** 

EXC -0.003159 0.000459 -6.883929 0.0000*** 

INT -0.014317 0.001712 -8.360204 0.0000*** 

C 0.864289 0.011510 75.08737 0.0000*** 

 

Period Fixed (Dummy Variables) 

R-squared 0.391486 Mean dependent var 0.866512 

Adjusted R-squared 0.288153 S.D. dependent var 0.358401 

S.E. of regression 0.302386 Akaike info criterion 0.590395 

Sum squared resid 4.846187 Schwarz criterion 0.930575 

 Log likelihood -8.597435 Hannan-Quinn criter. 0.724189 

F-statistic 3.788603 

 Prob(F-statistic) 0.000967*** 
 

 

As a result, two way fixed effects model is used to analyze the relationship between financial risks and market 

value of commercial banks. White panel corrected standard errors is used to cope with autocorrelation and 

heteroscedasticity problems. The results are shown in Table 11. The model is significant with 1% level of 

significance. R2 value of %28,81 means that financial risks (exchange rate risk, interest rate risk and liquidity risk) 



Financial Risk and Management Reviews, 2020, 6(1): 79-87 

 

 
86 

© 2020 Conscientia Beam. All Rights Reserved. 

affect around %29 of market price movement. Therefore, there are statistically significant relationships between 

exchange rate risk, interest rate risk and liquidity risk and market value. 

 

6. CONCLUSION 

Volatility in global markets and the pandemic increases the importance of banks. Countries’ stability and 

growth require strong commercial banks. In recent years, asset size and profitability of Turkish Banks have 

increased. Meanwhile, the stock prices of commercial banks haven’t increased that much on Istanbul Stock 

Exchange.  

The most important barriers of stock price increase are the financial risks. In this study, 7 commercial banks on 

Istanbul Stock Exchange are analyzed. Financial risks available in this study are interest rate risk, exchange rate 

risk and liquidity risk. The motivation of the study is to test the relationship between financial risks and market 

value. According to the results of the research, there is a statistically significant and positive relationship between 

liquidity position and firm value. A unit increase in liquidity position results in 5.79 units increase in firm value. 

Liquidity position is computed by dividing liquid assets to total assets. This result highlights the importance of 

holding liquid assets in commercial banks. Liquidity risk in general is related to the inability of paying back the 

current liabilities. For banks, this requires an additional maturity mismatch analysis. However, strong liquidity 

positions contribute to the market value of commercial banks.  

There is also statistically significant but negative relationship between firm value and exchange rate risk. A 

unit increase in exchange rate risk results in 0.003 unit decrease in firm value. Excessive use of speculative foreign 

exchange positions in bank treasuries may lead to decreases in their market value. Therefore the results of this 

study are also consistent with financial risk literature.  

On the other hand, there is a statistically significant and negative relationship between interest rate risk and 

firm value. Foreign exchange and interest rate is closely connected in finance theory, so this explains similar 

patterns that occur as a result of this study. The primary operations of banks; loans and deposits are interest 

related. Therefore interest rate is an important risk for commercial banks. In addition, central bank of Turkey, often 

changes the market interest rates in order to control the volatility in exchange rates. This results in losses for 

banks in wrong position.  

As a result, economic situation in Turkey combined with maturity mismatch contain threats for banks 

operating in Turkey. Since banks work under intense financial risks, these risks have to be measured, categorized 

and managed properly. This research shows that Banks don’t adequately manage financial risks. Increase in the use 

of hedging instruments such as derivatives, and corporate risk management would make the banks more resistant.  

 
Funding: This study received no specific financial support.    
Competing Interests: The authors declare that they have no competing interests.  
Acknowledgement: Both authors contributed equally to the conception and design of the 
study. 

 

REFERENCES 

Baltagi, B. H., & Li, Q. (1991). A joint test for serial correlation and random individual effects. Statistics & Probability Letters, 

11(3), 277–280. 

Born, B., & Breitung, J. (2016). Testing for serial correlation in fixed-effects panel data models. Econometric Reviews, 35(7), 1290-

1316. 

Chapman, R. J. (2006). Simple tools and techniques for enterprise risk management: John Wiley and Sons. 

Kocak, K. (2012). Mixed distribution model approach in financial risk analysis. Cukurova University Graduate School of Science, 

Master Thesis, Adana.    



Financial Risk and Management Reviews, 2020, 6(1): 79-87 

 

 
87 

© 2020 Conscientia Beam. All Rights Reserved. 

Kok, R., Ekinci, R., & Ay, Y. A. E. (2017). Country risk impact on corporate sector components: The case of Turkey and 

Azerbaijan-Kazakistanrusy. Bilig, 83, 281-302. 

Koroglu, Y. (2019). Mixed distribution model in financial risk management. Necmettin Erbakan University. Institute of Science. 

Master Thesis. Konya.    

Reis, S. G., Kilic, Y., & Bugan, M. F. (2016). Factors that affect bank profitability: The case of Turkey. The Journal of Accounting 

and Finance, 72, 21-36. 

Saldanli, A., & Aydin, M. (2016). Investigation of the factors affecting profitability in the banking sector with panel data analysis: 

The case of Turkey. Journal of Economics and Statistics, 24, 1-9. 

Senol, Z., & Karaca, S. S. (2017). The effect of enterprise risk management on firm performance: A case study on Turkey. 

Financial Studies, 21(2), 6-30. 

Senol, Z., Oncül, M., & Buyer, M. S. (2019). Effect of bank financial risks on bank profitability. International Journal of 

Management Education and Economic Perspectives, 7(2), 101-109. 

Topaloglu, E. E. (2018). Determination of the relationship between financial risks and firm value: An application on Istanbul 

stock exchange companies. Journal of Mehmet Akif Ersoy Economics and Administrative Faculty, 5(2), 287-301. 

Unal, O., & Altin, H. (2010). Analysis of the relationship between exchange rate risk and firm value in the Istanbul stock 

exchange automotive sector. Dumlupinar University Journal of Social Sciences, 26, 277-287. 

Uzak, G. (2019). Financial risk hedge accounting practices in the real sector. Bahcesehir University, Masters Thesis, Istanbul.    

Yanartaş, M. (2010). A model suggestion for determining the financial risks of firms. Kadir Has University, Institute of Social 

Sciences, Doctoral Dissertarion, Istanbul.    

Yucel, A. T., Mandacı, P. E., & Kurt, G. (2007). Financial risk management of enterprises and use of Turev product: An 

application in businesses in ISE 100 index. Accounting and Finance Magazine, 36, 1-9. 

 

 

 

 

 

 

 

 

 

 

 

 

Views and opinions expressed in this article are the views and opinions of the author(s), Financial Risk and Management Reviews shall not be responsible or 
answerable for any loss, damage or liability etc. caused in relation to/arising out of the use of the content. 

 


