Does Bank Type Affect the Profitability? An Econometric Research on Conventional and Islamic Banks

Transcription

1. Research Article / Araştırma Makalesi Does Bank Type Affect the Profitability? An Econometric Research on Conventional and Islamic Banks* Banka Tipi Kârlılığı Etkiler mi? Konvansiyonel ve İslami Bankalar Üzerine Ekonometrik Bir Araştırma Suna Akten Çürük** - Ümran Münire Kahraman*** - Mine Işık**** Abstract: As in all world, in Turkey banks have quite an active position within the national economy. Banking sector providing transfer of capital accumulation to real investments, in 2005, with the arrangement, included deposit banks and development and investment banks as well as participation banks. Thus, two types of banks with different philosophy and principles had the opportunity to operate subject to the same regulations in the same competitive environment. Profitability is a matter that needs to be carefully examined in order to sustain the life of the enterprises. On the other hand, banking is a sector that is worth analyzing with the profitability numbers announced even during the global crisis. The aim of this study is to determine the factors affecting the profitability of banks. In this study, the data of 3 participation banks and 16 conventional banks operating in Turkish Banking Sector and having regular data between 2006-2018, was used. Descriptive and inferential statistical methods were used in the study. Independent internal variables, bank type, asset size, credit risk, liquidity risk, operational efficiency, capital structure; independent external variables, inflation and GDP growth rate; dependent variables were taken as return on equity and return on assets. The number of horizontal section units of the panel data is and the number of time units obtained for each horizontal section unit is . The data set consists of a total of 988 observations. Stata package program was used for data analysis. The study contributes to the literature by distinguishing it from other studies conducted in terms of analysis period, variables and sample. First, the effect of banking type on profitability was examined. Hypothesis tests were tested by creating a panel data regression where return on assets is a dependent variable. * This study was presented at the ‘International Congress of Islamic Economy, Finance and Ethics, ISEFE'19’ in Istanbul on 22 December 2019. ** Dr. Öğr. Üyesi, Necmettin Erbakan Üniversitesi, Siyasal Bilgiler Fakültesi, İşletme Bölümü Asst. Prof. Dr, Necmettin Erbakan University, Faculty of Political Sciences, Department of Business 0000-0001-5887-4905 sakten@erbakan.edu.tr *** Dr. Öğr. Üyesi, Necmettin Erbakan Üniversitesi, Siyasal Bilgiler Fakültesi, İşletme Bölümü Asst. Prof. Dr, Necmettin Erbakan University, Faculty of Political Sciences, Department of Business 0000-0002-9840-0461 uteksen@gmail.com **** Doktora Öğrencisi, Necmettin Erbakan Üniversitesi, Siyasal Bilgiler Fakültesi, İktisat Bölümü PhD Candidate, Necmettin Erbakan University, Faculty of Political Sciences, Department of Economics 0000-0002-6363-7143 mineisiik@gmail.com Cite as/ Atıf: Akten Çürük, S., Kahraman, Ü. M., Işık, M. (2020). Makale başlığı. Turkish Studies - Economy, 15(2), 727-747. https://dx.doi.org/10.29228/TurkishStudies. Received/Geliş: 02 April/Nisan 2020 Checked by plagiarism software Accepted/Kabul: 20 June/Haziran 2020 Published/Yayın: 25 June/Haziran 2020 Copyright © INTAC LTD, Turkey CC BY-NC 4.0
2. 728 Suna Akten Çürük - Ümran Münire Kahraman - Mine Işık Similarly, the regression result was obtained for the return on equity variable. As a result, no significant impact was detected in models that tested the effect of bank type (conventional / Islamic) on return on equity and return on assets. In the second stage, In the second stage; in order to examine whether the variables differ according to bank type, t test was performed. As a result, it was found that operational efficiency, total assets representing the bank size, liquidity risk and financial risk were significantly lower in Islamic banks than in conventional banks. Structured Abstract: The purpose of the study to determine the effect of banking type on the level of profitability and which variables has an affect on the profitability of banks, which consists of 3 participation banks and 16 conventional banks from the two types of banks operating in Turkey for the period covering 2006-2018 years. Return on asset and return on equity variables is used as dependent variables in the analysis. While the bank type, asset size, credit risk, liquidity risk, financial risk and operational efficiency are used as independent internal variables, macroeconomic indicators representing by inflation and GDP growth rate are used as independent external variables. The number of horizontal section units of the panel data is n = 19 and the number of time units obtained for each horizontal section unit is t = 52. The data set consists of a total of 988 observations. Stata package program was used for data analysis. When the academic studies that comparatively examine the financial performances of Islamic banks and conventional banks is considered, different results can be observed. The study carried out to examine the effect of bank type on profitability differs from other studies conducted in terms of selected analysis period, sample and variables and makes a significant contribution to the literature. In the analysis, data for deposit banks got from the independent auditors' quarterly report issued at the official web page of the Banks Association of Turkey (TBB). Data for participation banks got from the independent auditors' quarterly report issued at the official web page of the Participation Banks Association of Turkey (TKBB). The inflation and growth rates data obtained from the official web page of Turkish Statistical Institute (TUİK). The monthly available data has been converted into quarterly data for 4 periods. In this study, descriptive statistics of the variables that are thought to affect the asset profitability of banks are determined, and the profitability variable trend is revealed. First, the effect of banking type on profitability was examined. Hypothesis tests were tested by creating a panel data regression where return on assets is a dependent variable. Similarly, the regression result was obtained for the return on equity variable. In order to see whether classical banking and Islamic banking systems have an impact on profitability, a dummy variable that indicates the banking system is included as a descriptive variable in the regression model. In the second stage, t-test was applied to determine the variables that differ according to the bank type. As a result of the analysis; it is concluded that to made a classification as Islamic banks and conventional banks does not have an impact on profitability. Even if the working principles and philosophy are different, the similarity of the functioning structure of the two bank types and their operation in the same competitive environment and subject to the same legal regulations can be considered as one of the main reasons of this result. According to the results of the t-test applied to examine whether the effects of the variables differ according to the bank type; it is determined that operational efficiency, bank size, liquidity risk and financial risk were significantly lower in Islamic banks than conventional banks. Determining that operational efficiency is significantly lower in Islamic banks than conventional banks shows that the share of Operating Expenses (Dividend Expenses + Commercial Loss) in Net Dividend Revenues is lower in Islamic banks. The fact that the asset size value representing the bank size is lower in Islamic banks compared to conventional banks supports the fact that the share of Islamic banks in the banking sector is below the target. This result coincides with the understanding that Islamic banks bring the funds collected by working with the principle of profit and risk sharing to the economy, but not aim the profit maximization. Liquidity status is related with banks' fulfilling their deposit, debt and credit obligations. Determining that the liquidity risk is significantly lower in Islamic banks than conventional banks; shows that the total debt divided by the total deposit ratio is lower in islamic banks. This result supports that the performance of Islamic banks yielded better results than conventional banks during the 2008 Global Financial Crisis period. The result Turkish Studies - Economy, 15(2)
3. Does Bank Type Affect the Profitability ? An Econometric Research on Conventional… 729 that equity capital has less share in the capital structure of Islamic banks than conventional banks means the financial risk level is significantly lower in Islamic banks. Developing unique and modern products in accordance with Islamic finance understanding will enable to reveal their differences within the sector. In addition, Islamic banks are expected to increase permanently their shares in the sector by paying attention to Islamic financial literacy in order to ensure the public understanding of the financial products and services that they use. So, they will be able to compete more successfully on the international platforms. Keywords: Islamic Finance, Islamic Banking, Conventional Banking, Financial Performance, Panel Data Analysis Jel Codes: G2, G21 Öz: Tüm dünyada olduğu gibi Türkiye’de bankalar ülke ekonomisi içinde oldukça etkin bir konuma sahiptir. Sermaye birikiminin reel yatırımlara aktarılmasını sağlayan bankacılık sektörü 2005 yılında yapılan düzenlemeyle birlikte, mevduat bankaları, kalkınma ve yatırım bankalarının yanı sıra katılım bankalarını da içine almıştır. Böylece felsefesi ve ilkeleri farklı olan iki banka tipi, aynı rekabet ortamında aynı düzenlemelere tâbi olarak faaliyet gösterme imkânı bulmuştur. Kârlılık, işletmelerin hayatını devam ettirebilmesi açısından dikkatle incelenmesi gereken bir meseledir. Bankacılık ise küresel kriz sürecinde bile açıkladığı kârlılık rakamları ile analiz edilmeye değer bir sektördür. Bu çalışmanın amacı, bankaların karlılığını etkileyen faktörlerin belirlenmesidir. Çalışmada, Türk Bankacılık Sektöründe faaliyet gösteren ve 2006-2018 yılları arasında düzenli veriye sahip 3 katılım bankası ile 16 konvansiyonel banka verileri kullanılmıştır. Çalışmada betimsel ve çıkarımsal istatistik yöntemlerinden faydalanılmıştır. Bağımsız iç değişkenler, banka tipi, aktif büyüklüğü, kredi riski, likidite riski, faaliyet verimliliği, sermaye yapısı; bağımsız dış değişkenler, enflasyon ve GSMH büyüme oranı; bağımlı değişkenler ise öz kaynak karlılığı ve aktif karlılığı olarak alınmıştır. Panel verinin yatay kesit birim sayısı , her bir yatay kesit birimi için elde edilen zaman birimi sayısı ise ’dir. Veri seti toplam 988 gözlemden oluşmaktadır. Veri çözümlemesinde Stata paket programından faydalanılmıştır. Çalışma analiz dönemi, değişkenler ve örneklem bakımından yapılmış olan diğer çalışmalardan ayrışarak, literatüre katkı sağlamaktadır. İlk olarak, bankacılık tipinin kârlılık üzerindeki etkisi incelenmiştir. Aktif kârlılığın bağımlı değişken olduğu bir panel veri regresyonu oluşturularak hipotez testleri sınanmıştır. Benzer şekilde özsermaye getiri oranı değişkeni için de regresyon sonucu elde edilmiştir. Sonuçta, banka tipinin (konvansiyonel/katılım) öz kaynak karlılığı ve aktif karlılığı üzerindeki etkisini test eden modellerde, önemli düzeyde bir etki tespit edilememiştir. İkinci aşamada; banka tipine göre değişkenlerin farklılık gösterip göstermediğini incelemek amacıyla t testi yapılmıştır. Sonuçta faaliyet verimliliğinin, banka büyüklüğünü temsil eden aktif toplamının, likidite riskinin ve finansal riskin, konvansiyonel bankalara göre katılım bankalarında anlamlı düzeyde daha düşük olduğu sonucu elde edilmiştir. Anahtar Kelimeler: İslami Finans, İslami Bankacılık, Konvansiyonel Bankacılık, Finansal Performans, Panel Veri Analizi Jel Kodları: G2, G21 Introduction As in all around the world, banks have quite an active position within the national economy in Turkey. This effective position of banks in the country's economies stems from the task of bringing together those has surplus fund with those need fund. It is very necessary to ensure high performance in the banking sector. The most basic reason behind this is that the effects of a possible trouble that may occur in the banking sector will reflect on the entire economy. Therefore, countries carry out many studies to make the banking sector more effective and academic studies examining the profitability of banks are also important. In this context, the subject of the study consists of examining the effect of bank type (Islamic-conventional) on profitability and determining the variables that affect the profitability of banks. www.turkishstudies.net/economy
4. 730 Suna Akten Çürük - Ümran Münire Kahraman - Mine Işık The banking sector that enables the transfer of capital accumulation to real investments also included participation banks as well as deposit banks and development banks with an arrangement made in 2005 in Turkey. Participation banks which are getting more important in the financial system got start with the name of “private financial institutions” in Turkey in 1983. The purpose of the establishment of this institutions is to bring the savings of people with interest sensitivity to the economy and to ensure that similar foreign funds come to the country. The private finance institutions has been named as “Participation Bank” since 2005 (Parlakkaya & Akten Çürük, 2011: 37). So, two types of banks, whose philosophy and principles are different, had the opportunity to operate under the same regulations in the same competitive environment. The aim of the study, which consists of 3 participation banks and 16 conventional banks from the two types of banks operating in Turkey, to determine the effect on the level of profitability and to determine the variables that affect the profitability of banks for the period covering the period of 2006-2018. For this purpose, firstly, panel data regression analysis was performed to determine the effect of banking type on profitability. In the second stage, t-test was performed to determine the variables that differ according to the bank type. It is thought that the study will make a significant contribution to the literature by distinguishing it from other studies conducted in terms of selected analysis period, variables and sampling. The study consists of four chapters. After the introduction, a literature review is given in the first part. After the second section, which includes the methodology of the study, the empirical findings obtained were reported and interpreted in the third section. The study has been completed with the fourth section where a general evaluation is made. Literature Review A large number of academic studies was examined, which comparatively analyzed the financial performance of Islamic banks and conventional banks (Table 1). It was observed that different results were detected in these studies. When the following literature is examined, which consists of two parts in the form of studies conducted in Turkey and abroad. (Iqbal, 2001), (Jaffar & Manarvi, 2011), (Usman & Khan, 2012), (Siraj & Pillai, 2012), (Parlakkaya & Akten Çürük, 2011), (Şamiloğlu et al. 2013) are concluded that the performance of Islamic banks is higher than the conventional banks. On the other hand, (Safiullah, 2010), (Fayed, 2013), (Toraman et al. 2015), (Tuncay, 2018) have concluded that conventional banks are more effective than Islamic banks. (Samad & Hasan, The performance of Malaysian Islamic bank during 1984-1997: An exploratory study, 1999), (Samad, 2004) and (Loghod, 2005), (Sakarya & Kaya, 2013), (Doğan, 2013) et al. found that there is no difference between financial performances of two type banks. All these results caused debates and caused an increase in the academic studies on this subject. Table 1: Related Studies Author(s) Method Samad and Hasan (1999) Ratio Analysis Iqbal (2001) Ratio Analysis Studies Conducted Abroad Time Extent Period Result 19841997 Malaysia (8 Islamic and 8 Conventional Banks) It is determined that the bank type does not differ statistically in terms of profitability. 1990-998 12 Islamic and Conventional Banks operating in 9 Countries It is determined that Islamic banks outperform conventional banks in terms of liquidity and profitability almost every year. Turkish Studies - Economy, 15(2)
5. Does Bank Type Affect the Profitability ? An Econometric Research on Conventional… 731 Bahrain (Islamic and Conventional Banks) It is concluded that the bank type does not differ significantly in terms of profitability and liquidity performance. Islamic and Conventional Banks operating in 5 Countries Bangladesh (4 Islamic Banks and 4 Conventional Banks) It has been determined that there is no significant difference between the performances of Islamic and conventional banks. It has been determined that the performances of conventional banks are better than Islamic banks. 20052009 Pakistan (5 Islamic Banks and 5 Conventional Banks) It is determined that Islamic banks outperform conventional banks and provide liquidity. T-test ANOVA 20062009 Pakistan (3 Islamic Banks and 3 Conventional Banks) There was no statistically significant difference between Islamic and conventional banks in terms of profitability performance. ANOVA 20052010 Gulf Region (6 Islamic Banks and 6 Conventional Banks) It is determined that Islamic banks have higher rate of return on equity and rate of return on assets than conventional banks. MANOVA 20072009 46 Islamic Banks and 49 Conventional Banks from 15 Countries It was determined that Islamic banks performed better than conventional banks in 2007, and that conventional banks performed better in 2009. Usman and Khan (2012) T-test 20072009 Pakistan (3 Islamic Banks and 3 Conventional Banks) It is determined that Islamic banks have high liquidity power, high growth and profitability compared to conventional banks. Fayed (2013) Ratio Analysis 20082010 Egypt (3 Islamic Banks and 6 Conventional Banks) It has been determined that conventional banks outperform Islamic banks. Ratio Analysis 19912001 Loghod (2005) Logit 20002005 Safiullah (2010) T-test 20042008 CAMEL Samad (2004) Jaffar and Manrvi (2011) Ansari and Rehman (2011) Siraj and Pillai (2012) Rashwan (2012) Author(s) Parlakkaya and Çürük (2011) Method T-test Logit Studies Conducted in Turkey Time Extent Period Turkey (4 Participation 2005Banks and 24 2008 Conventional Banks) www.turkishstudies.net/economy Result It has been determined that participation banks have higher profit margins than conventional banks.
6. 732 Suna Akten Çürük - Ümran Münire Kahraman - Mine Işık It is concluded that according to the total efficiency scores participation banks are more efficient than conventional banks in the period. It was found that participation banks didn’t differ significantly from conventional banks in terms of efficiency and profitability. Er and Uysal (2012) DEA 20052010 Turkey (4 Participation Banks and 26 Conventional Banks) Sakarya and Kaya (2013) Panel Data Analysis 20052012 Turkey (Participation Banks and Conventional Banks) TOPSIS 20032011 Turkey (4 Participation Banks and 4 Conventional Banks) Doğan (2013) T-test 20052011 Turkey (4 Participation Banks and 4 Conventional Banks) It has been determined that participation banks are more profitable and have higher performances than conventional banks. It was determined no statistically significant difference in the profitability of participation banks and conventional banks. Toraman et al. (2015) T-test GMM 20062014 Turkey (4 Participation Banks and 11 Conventional Banks) It has been determined that the total assets and liquid assets of conventional banks are higher than those of participation banks. DEA 20062014 Yurttadur and Demirbaş (2017) T-test 20102016 Tuncay (2018) T-test Mann Whitney-U test Şamiloğlu et al. (2013) Ata and Buğan (2016) Year 2016 Turkey (4 Participation Banks and 11 Conventional Banks) Turkey (3 Participation Banks and 3 Conventional Banks) Turkey (5 Participation Banks and 17 Conventional Banks) Participation banks were found to be more efficient than conventional banks. Participation banks were found to perform better in terms of profitability than conventional banks. Although not statistically significant, profitability was found to outperform conventional banks than participation banks. According to the literature examined, of this study; it is observed that it differs from other studies conducted in terms of analysis period, variables and selected sample. Thus, the study will make an significant contribution to the literature. Methodology of the Research Purpose, Extent and Importance of the Research The purpose of this study, to determine the effect on the profitability levels of the two types of banks and to identify variables that affect the profitability of banks, for the period covering the years 2006-2018 in Turkey. The extent of this study; although their philosophy and principles are different, they constitute two types of banks consisting of 3 participation and 16 conventional banks operating under the same regulations in the same competitive environment. For this aim, firstly, panel data regression analysis was performed to determine the effect of banking type on profitability. at the second stage, t-test was performed to determine the variables that differ according to the bank type. The study will make a significant contribution to the literature by distinguishing it from other studies conducted in terms of selected analysis period, variables and sampling. Turkish Studies - Economy, 15(2)
7. Does Bank Type Affect the Profitability ? An Econometric Research on Conventional… 733 The banking sector is a very important sector that is worth analyzing with its profitability figures announced even during the global crisis. Countries are carrying out many studies to make the banking sector work more effectively. Therefore, academic studies examining the profitability of banks are very important. In this context, the study carried out to examine the effect of bank type on profitability differs from other studies conducted in terms of selected analysis period, sample and variables and makes a significant contribution to the literature. Data Set and Sample of the Research In this study, we used data from quarterly financal tables of 3 participation banks and 16 conventional banks, which operate in the Turkish banking sector and have regular data between 2006-2018. The bank names included in the analysis are given in Table 2. Table 2: Banks Included in the Analysis Conventional Banks (Deposit Banks) Public Capital Deposit Banks Banks Funded by Foreign Capital 1 Türkiye Cumhuriyeti Ziraat Bank A.Ş. 2 Türkiye Halk Bank A.Ş. 3 Türkiye Vakıflar Bank T.A.O. Private Equity Deposit Banks 4 Akbank T.A.Ş. 5 Anadolubank A.Ş. 6 Şekerbank T.A.Ş. 7 Turkish Bank A.Ş. 8 Türk Ekonomi Bank A.Ş. 9 Türkiye İş Bank A.Ş. 11 Alternatifbank A.Ş. 12 Citibank A.Ş. 13 Denizbank A.Ş. 14 HSBC Bank A.Ş. 15 QNB Finansbank A.Ş. 16 Türkiye Garanti Bank A.Ş. 10 Yapı ve Kredi Bank A.Ş. Participation Banks (Islamic Banks) 1 Albaraka Türk Participation Banks A.Ş. Horizontal Section Units n=19 2 Kuveyt Türk Participation Banks A.Ş. 3 Türkiye Finans Participation Banks A.Ş. Number of Time Units t=52 Total Observations = 988 Three of conventional banks are public banks, six of them are foreign banks and seven of them are private banks. Only 3 participation banks from 6 participation banks currently operating in Turkey is included in he analysis. In this case, the number of horizontal section units representing the number of banks in the analysis is N = 19, and the number of time units obtained for each horizontal section unit is T = 52. Therefore, the data set consists of a total of 988 observations. The rate of return on assets and the rate of return on equity are used as dependent variables. Bank type, asset size, credit risk, liquidity risk, financial risk and activity efficiency represents independent internal variables. As macroeconomic indicators representing independent external variables in the analysis; inflation and GDP growth rate is used. Abbreviations and calculations of the variables used are given in Table 3. Table 3: Variables Used in the Analysis Variable Name Used Return on Assets Return on Equity Acronym ROA ROE Calculation of Variables Net Profit / Total Asset Net Profit / Equity Total www.turkishstudies.net/economy
8. 734 Suna Akten Çürük - Ümran Münire Kahraman - Mine Işık Activity Efficiency AE Asset Size Credit Risk Liquidity Risk Financial Risk Inflation GNP Growth Rate AS CR LR FR INF GNP Operating Expenses / Net Interest (Dividend) Income Logarithm of Assets Non-Performing Loans / Total Loans Total Debt / Total Deposits Equity Total / Total Assets Data Collection Technique and Tool of the Research In the analysis, data for deposit banks got from the independent auditors' quarterly report issued at the official web page of the Banks Association of Turkey (TBB). Data for participation banks got from the independent auditors' quarterly report issued at the official web page of the Participation Banks Association of Turkey (TKBB). The inflation and growth rates data obtained from the official web page of Turkish Statistical Institute (TUIK). The monthly available data has been converted into quarterly data for 4 periods. Research Method In this study, descriptive statistics of the variables that are thought to affect the asset profitability of banks are determined, and the profitability variable trend is revealed. First, the effect of banking type on profitability was examined. Hypothesis tests were tested by creating a panel data regression where return on assets is a dependent variable. Similarly, the regression result was obtained for the return on equity variable. In order to see whether classical banking and Islamic banking systems have an impact on profitability, a dummy variable that indicates the banking system is included as an independent variable in the regression model. In order to see the effect of dummy variable in the model, pooled OLS model is used. In the second stage, t-test was applied to determine the variables that differ according to the bank type. Stata package program was used in data analysis. To determine the independent variables that affect the profitability of banks; Ø :Explanatory variables have no effect on asset profitability. :Explanatory variables have an impact on asset profitability. Ø :Explanatory variables have no effect on equity profitability. :Explanatory variables have an impact on equity profitability. in this way alternative hypotheses has been formed. In order to determine whether of two types banks consisting of conventional and Islamic banks affect profitability; Ø :Banking type does not affect asset profitability. Turkish Studies - Economy, 15(2)
9. Does Bank Type Affect the Profitability ? An Econometric Research on Conventional… 735 :Banking type affects asset profitability. :Banking type does not affect equity profitability. Ø :Banking type affects equity profitability. in this way hypotheses has been formed. Panel data regression equations created to test (1) and (2). and hypotheses are given in equations (1) (2) , In Equation 1, created to test the hypothesis, while the asset profitability also expressed as the Rate of Return on Assets (ROA), represents the dependent variable; activity efficiency (AE), asset size (AS), credit risk (CR), liquidity risk (LR), financial risk (FR) independent internal variables, inflation (INF) and GNP growth rate represent independent external variables. In the 2nd Equation created to test the hypothesis; equity profitability also referred to as Rate Return on Equity (ROE), represents the dependent variable. To test the and hypotheses, dummy variables were added to equations (1) and (2). (3) www.turkishstudies.net/economy
10. 736 Suna Akten Çürük - Ümran Münire Kahraman - Mine Işık (4) , in equations (3) and (4); If the bank is a classic bank it takes value 1, in other case it takes value 0. On the other hand in equations (3) and (4); If the bank is a Islamic bank it takes the value 1, in other case it takes value 0. It refers to ‘ ’ bank number, ‘ ’ time units, ‘ ’ error terms, ‘ ’ constant terms and regression coefficients in the equations. Empirical Findings Obtained in the Research Descriptive statistical information about the variables used in the analysis; mean, standard deviation, number of observations, minimum and maximum values are given in Table 4. Descriptive statistics related to variables were calculated separately for general, horizontal section units and time period. Table 4: Descriptive Statistics of Variables Variable ROA ROE AE AS CR LR FR INF Overall Within Between Overall Within Between Overall Within Between Overall Within Between Overall Within Between Overall Within Between Overall Within Between Overall Mean 0.0103 0.0919 1.2523 7.4737 0.0701 1.4040 0.1127 9.0928 Std. Deviation 0.0071 0.0029 0.0065 0.0619 0.0282 0.0555 0.4850 0.1985 0.4448 0.6789 0.6242 0.3024 0.9407 0.1317 0.9319 0.1830 0.1215 0.1396 0.0253 0.0181 0.0182 2.9716 Min. -0.0146 0.0029 -0.0112 -0.1566 0.0185 -0.1109 0.0787 0.7943 0.2536 5.7404 6.0048 6.6659 0.0000 0.0190 -0.5181 1.0313 1.1926 1.0079 0.0568 0.0888 0.0508 4.3500 Turkish Studies - Economy, 15(2) Max. 0.0486 0.0140 0.0460 0.3527 0.1438 0.3382 6.7031 1.6488 6.8779 8.7419 8.2488 8.1739 29.5995 0.6105 29.0591 2.1444 1.6302 2.3387 0.2142 0.1549 0.1728 22.3867 Observations 988 N = 19 T = 52 988 N = 19 T = 52 988 N = 19 T = 52 988 N = 19 T = 52 988 N = 19 T = 52 988 N = 19 T = 52 988 N = 19 T = 52 988
11. Does Bank Type Affect the Profitability ? An Econometric Research on Conventional… GNP Within Between Overall Within Between 5.1154 0.0000 2.9716 5.0056 0.0000 5.0056 9.0928 4.3500 -14.6000 5.1154 -14.6000 9.0928 22.3867 12.6000 5.1154 12.6000 737 N = 19 T = 52 988 N = 19 T = 52 The trend of asset profitability by years for the banks is given in Table 5. Table 5: Average Values of ROA by Years Years Banking Sector Conventional Banks Islamic Banks 2006 0.0135 0.0131 0.0153 2007 0.0157 0.0158 0.015 2008 0.013 0.0127 0.015 2009 0.0143 0.0147 0.0143 2010 0.0126 0.0127 0.0118 2011 0.0085 0.0083 0.0097 2012 0.01 0.0099 0.0103 2013 0.0085 0.0086 0.0084 2014 0.0073 0.0074 0.007 2015 0.006 0.0061 0.0056 2016 0.0071 0.0073 0.0058 2017 0.0082 0.0087 0.006 2018 0.0085 0.009 0.0061 The trend of return on assets as of the years in which the sector means and conventional and Islamic banks averages are exhibited, with the graph in Figure 1. www.turkishstudies.net/economy
12. 738 Suna Akten Çürük - Ümran Münire Kahraman - Mine Işık 0,018 0,016 0,014 0,012 0,01 0,008 0,006 0,004 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Banking Sector Conventional Banks Islamic Banks Figure 1: Trend and Comparision of ROA by Years Looking at the graph, it is seen that Islamic banks displayed a higher return on assets than conventional banks with a noticeable difference in 2006 and 2008. Despite the global financial crisis in 2008, the high asset profitability in Islamic banks has increased the interest and trust in Islamic banks after this year. It is observed that the average of active profitability, which ran parallel to each other in 2009 and 2010, was higher in Islamic banks than in conventional banks in 2011. It is observed that the average of active profitability, which remained parallel to each other from 2012 to 2015, was higher in conventional banks than in Islamic banks after 2015. Table 6 shows the summary of annual of the change in asset profitability of the sector with conventional banking and Islamic banking. Table 6: Summary of Annual ROA Banking Sector 2006 0.0131 0.0153 -0.0012 0.0417 0.0135 0.0090 Coefficient of Variation 0.6674 2007 0.0158 0.015 0.0020 0.0395 0.0157 0.0087 0.5561 2008 0.0127 0.015 0.0010 0.0284 0.013 0.0056 0.4281 2009 0.0147 0.0143 0.0011 0.0314 0.0143 0.0071 0.4975 2010 0.0127 0.0118 0.0017 0.0276 0.0126 0.0068 0.5422 2011 0.0083 0.0097 -0.0006 0.0243 0.0085 0.0060 0.7027 2012 0.0099 0.0103 0.0020 0.01 0.0087 0.8731 2013 0.0086 0.0084 -0.0020 0.0215 0.0085 0.0050 0.5865 2014 0.0074 0.007 -0.0017 0.0203 0.0073 0.0049 0.6729 2015 0.0061 0.0056 -0.0105 0.0205 0.006 0.0049 0.8092 2016 0.0073 0.0058 -0.0146 0.0241 0.0071 0.0064 0.9001 2017 0.0087 0.006 0.0003 0.0082 0.0055 0.6755 Conventional Islamic Banks Banks Min. Max. Average 0.0395 0.0315 Turkish Studies - Economy, 15(2) Std. Deviation
13. Does Bank Type Affect the Profitability ? An Econometric Research on Conventional… 2018 0.009 0.0061 0.0009 0.0486 0.0085 Average Std. Deviation Coefficient of Variation 0.0103 0.0100 0.0102 0.0031 0.0039 0.0032 0.3001 0.3862 0.3085 0.0068 739 0.7998 We can analyze the risk situation by looking at the table containing the average, standard deviation and coefficient of variation accepted as risk indicators. When the table is analyzed, it is seen that the average of return on assets is close to conventional banking and Islamic banking. When these values are compared with the general average of the banking sector, it is seen that the values are close to the general average of the banking sector. Investigation of Multicollinearity Independent variables must be unrelated to perform regression analysis. For this reason, the correlation matrix, which gives an idea about the relationship of variables, is examined and the results are presented in Table 7. Table 7: Correlation Coefficients for Independent Variables AE AE AS CR LR FR INF GNP 1 0.006 -0.014 -0.057 -0.293 0.270 -0.145 1 0.002 0.268 -0.349 0.130 0.042 1 0.003 0.007 0.014 0.008 1 -0.211 0.008 0.075 1 -0.153 -0.040 1 -0.193 AS CR LR FR INF GNP 1 When correlations between independent variables are evaluated in Table 6, it is observed that there is no significant correlation. In Table 8, tolerance and Variance Inflation Factors (VIF) values were obtained to investigate the multicollinearity problem (Alpar, 2013). When the findings obtained were evaluated in general, it was determined that there were no multicollinearity problems between the independent variables. Table 8: Tolerance and VIF Values for Independent Variables Variables Tolerance VIF AE 0.830 1.204 www.turkishstudies.net/economy
14. 740 Suna Akten Çürük - Ümran Münire Kahraman - Mine Işık AS CR 0.811 0.999 1.233 1.001 LR FR 0.908 0.749 1.102 1.335 INF GNP 0.878 0.941 1.139 1.062 Panel Data Regression Analysis Panel data regression analysis method is suitable when both cross section and time dimension are available in the data. Since the data used in the research are quarterly, seasonal dummy variables are added to the regression equation and season effects are eliminated. Firstly, it is examined whether there is a cross-section dependency. If there is a cross-section dependency between the units, the analysis results can be deviated and inconsistent (Topaloğlu, 2017). LMadj test of Pesaran et al. (2008) is used in cases where the T dimension is larger than the N dimension. For the LMadj test, the H0 hypothesis is created as no cross-sectional dependency. When applying the LMadj test, the dependent variable was calculated separately for ROA and ROE. Also, INF and GNP variables are not included in the test since they are the same in all cross sections. ( ( section dependency. ) was obtained for the ROA variable. For the ROE variable, . Error terms of the regression model built with variables have cross- Unit root analysis was performed in panel data variables due to the length of time. Since cross-section dependency is detected in the data, cross-section extended CIPS statistics were used developed by Im et al. (2003) (Yerdelen Tatoğlu, 2018). The stationary of the variables of INF and GNP was investigated with the PP statistics of Phillips and Perron (1988). For both tests, the H0 hypothesis states that the series is not stationary. Results for unit root tests are given in Table 9. Maximum lag is taken as 2. When Table 9 is examined, the H0 hypothesis, which indicates that the CR, Inf and GNP variables contain unit root, could not be rejected for all lags. The series should be made stationary by taking the difference. Table 9: Unit Root Tests CIPS Lag 0 ROA Statistics -4.902 p value 0* ROE Statistics -5.097 p value 0* AE Statistics -2.961 p value 0.002* AS Statistics -3.29 p value 0.001* CR Statistics 0.174 p value 0.569 LR Statistics -5.982 p value 0* Constant 1 2 -5.379 -6.262 0* 0* -6.664 -6.977 0* 0* -8.309 -5.217 0* 0* -2.451 -2.785 0.007* 0.003* -0.067 -1.812 0.473 0.035** -3.268 -2.206 0.001* 0.014** Constant and Trend 0 1 2 -3.953 -4.977 -5.405 0* 0* 0* -4.271 -6.419 -7.066 0* 0* 0* -2.536 -10.25 -6.057 0.006* 0* 0* -3.519 -3.586 -4.732 0* 0* 0* 3.128 2.361 -0.016 0.999 0.991 0.494 -7.354 -4.927 -1.964 0* 0* 0.025** Turkish Studies - Economy, 15(2)
15. Does Bank Type Affect the Profitability ? An Econometric Research on Conventional… FR Statistics -2.017 -1.83 -1.619 p value 0.022** 0.045* 0.048** PP Constant Lag 0 1 2 INF Z(rho) 2.207 2.114 3.22 Z(t) 0.275 0.262 0.422 p value 0.9762 0.9755 0.9823 GNP Z(rho) -11.521 -11.574 -11.213 Z(t) -2.654 -2.659 -2.629 p value 0.0823 0.0815 0.0872 *: Significant at the 5% level **: Significant at the 1% level 741 -3.282 -1.826 -1.642 0.001* 0.034** 0.045** Constant and trend 0 1 2 0.76 2.665 6.142 0.107 0.44 1.162 0.9952 0.9967 1 -11.768 -11.811 -11.419 -2.424 -2.428 -2.388 0.3671 0.3648 0.3858 The unit root tests of the series in which the difference is taken is given in Table 10. While the CR and GNP variables become stationary with the first difference, the INF variable has become stationary by taking the second difference. In regression analysis, delayed values of the variables were used. Table 10: Variables Made Stable by Taking the Difference CIPS CR(1) Lag Statistics p value 0 -15.535 0** PP Lag 0 INF(2) Z(rho) -14.236 Z(t) -4.33 p value 0.0004** GNP(1) Z(rho) -15.428 Z(t) -3.967 p value 0.0016** *: Significant at the 5% level **: Significant at the 1% level Constant 1 -6.836 0** Constant 1 -14.193 -4.336 0.0004** -14.875 -3.992 0.0015** 2 -4.97 0** 2 -13.162 -4.573 0.0001** -13.684 -4.096 0.001** Constant and Trend 0 1 2 -14.247 -4.991 -2.863 0** 0** 0** Constant and Trend 0 1 2 -14.823 -14.735 -13.654 -5.003 -5.04 -5.757 0.0002** 0.0002** 0** -15.268 -14.176 -12.643 -3.792 -3.849 -4.066 0.0169* 0.0142* 0.0071** Models were estimated by pooled least squares (POLS) method to test H1, H2, H3 and H4 hypotheses. Model's whether errors are not autocorrelated and there is distortion in fixed variance assumptions are examined and the results are given in Table 11. Table 11: POLS Assumptions in Panel Data Dependent Variable H0 Test Test Statistics p Value Decision ROA No autocorrelation Wooldridge 169.577 0.000 H0 rejection www.turkishstudies.net/economy
16. 742 Suna Akten Çürük - Ümran Münire Kahraman - Mine Işık Fixed variance Breusch-Pagan / CookWeisberg Test 19.24 0.000 H0 rejection No autocorrelation Wooldridge 131.814 0.000 H0 rejection Fixed variance Breusch-Pagan / CookWeisberg Test 14.57 0.000 H0 rejection ROE The error terms of the model obtained by POLS method are assumed to be unautocorrelated and fixed variance. With the (Breusch & Pagan, 1979)/ (Cook & Weisberg, 1983) test, a regression model is established with independent variables by e taking dependent variable the squares of the error terms obtained from the POLS regression. The values obtained from the model are compared with the values. In this place, N is the number of observations and p is the number of independent variables. Wooldridge test is a robust method that can be applied in the absence of unit effects in the data to examine whether the error terms are autocorrelated (Drukker, 2003). in the model, the regression of error terms with lagged values is calculated. The validity of the regression is examined with the F test. According to Table 11, there is a deterioration in the assumptions. Therefore, the DriscollKraay estimator which produces robust standard errors, was used in the variance estimation of the parameters (Hoechle, 2007). The findings of parameter estimates are given in Table 12 for H1 and H3. It is given in Table 13 for H2 and H4. Table 12: Driscoll-Kraay Standard Errors POLS Coefficients for the ROA Dependent Variable H1 H3 ROA Coefficient Std. Error t P>t AE 0.000837 0.001033 0.81 0.428 AS 0.002505 0.000652 3.84 0.001* CR 1.62E-05 7.72E-05 0.21 0.836 LR -0.00947 0.001174 -8.07 0.000* FR 0.078759 0.013001 6.06 0.000* Inf 2.75E-05 0.000102 0.27 0.792 GNP -2.6E-05 0.000143 -0.18 0.86 Constant -0.00018 0.004896 -0.04 0.971 AE 0.00096 0.001039 0.92 0.368 AS 0.002573 0.000675 3.81 0.001* CR 1.66E-05 7.79E-05 0.21 0.834 LR -0.00918 0.001318 -6.97 0.000* FR 0.081896 0.014306 5.72 0.000* Inf 2.91E-05 0.000101 0.29 0.778 Turkish Studies - Economy, 15(2) R2 F P>F 0.46 51.31 0.000* 0.46 109.27 0.000*
17. Does Bank Type Affect the Profitability ? An Econometric Research on Conventional… GNP -2.6E-05 0.000143 -0.18 0.86 Part. 0.000472 0.000642 0.74 0.471 Constant -0.00169 *: Significant at the 5% level 0.006102 -0.28 0.785 743 In Table 12, where the results for the H1 hypothesis, where the rate of return on assets (ROA) is a dependent variable, the asset size has a positive effect (0.00251) on asset profitability. Liquidity risk has a decreasing effect of -0.009 units on active profitability. The impact of financial risk on asset profitability is positive and is 0.0788 units. In the model obtained for the H3 hypothesis created to test the impact of the banking system on asset profitability, no significant impact on profitability has been identified in the variation of the bank type as Islamic bank or conventional bank. Table 13: Driscoll-Kraay Standard Errors POLS Coefficients for the ROE Dependent Variable H2 ROE Coefficient Std. Error t P>t AE AS 0.011981 0.009781 1.22 0.236 0.019667 0.005899 3.33 0.004* CR 0.000193 0.00067 0.29 0.777 LR FR Inf -0.09417 -0.04305 0.011738 0.109564 -8.02 -0.39 0.000* 0.699 0.000297 0.000903 0.33 0.746 GNP -0.00018 0.001244 -0.15 0.886 Constant 0.110293 0.012075 0.04915 0.009604 2.24 1.26 0.038 0.225 0.01972 0.006249 3.16 0.005* CR 0.000193 0.000671 0.29 0.777 LR AE AS H4 FR Inf -0.09394 -0.04063 0.01306 0.13912 -7.19 -0.29 0.000* 0.774 0.000299 0.000892 0.33 0.742 GNP -0.00018 0.001244 -0.15 0.886 Part. 0.000364 0.109131 0.006985 0.063413 0.05 1.72 0.959 0.102 Constant R2 F P>F 0.44 70.76 0.000* 0.44 116.74 0.000* *: Significant at the 5% level In Table 13, where the results for the H2 hypothesis, where the rate of return on equity (ROE) is dependent variable, it is seen that the size of the bank has a positive effect on the return on equity (0.01967). Liquidity risk has a decreasing effect of -0.094 units on equity profitability. The H4 hypothesis was tested to see the effect of the banking system on equity profitability. According to the findings, there was no significant effect on the profitability of the bank type varying as Islamic bank or conventional bank. t-Test To examine whether the variables differ according to the bank type; Ø :It does not differ according to the variable banking type. www.turkishstudies.net/economy
18. 744 Suna Akten Çürük - Ümran Münire Kahraman - Mine Işık :It varies according to the variable banking type. alternative hypotheses has been formed. Table 14: Variables that Differ Between Islamic Banks and Conventional Banks Variable t p value Mean difference AE -2.705 0.007* -0.11442 AS -8.584 0.000* -0.34969 CR -.524 0.601 -0.04322 LR -9.867 0.000* -0.15818 FR -7.648 0.000* -0.01409 INF 0.002 0.998 0.00055 GNP 0.015 0.988 0.00639 ROA -0.880 0.380 -0.00049 ROE 1.019 0.309 0.00550 *: Significant at the 5% level The t-test results applied to the hypotheses formed to determine whether the variables differ by bank type are reported in Table 14. The data Islamic bank was prepared as 1 and the conventional bank as 2. Findings obtained in Table 14; activity efficiency, asset size (total assets), liquidity risk and financial risk variables are significantly lower in Islamic banks than in conventional banks. Conclusion In this study, it is aimed to investigate the effects of bank type on profitability. In this context, panel data regression analysis is carried out by using data from quarterly financial tables of 16 conventional banks and 3 participation banks, which had regular data between 2006-2018. In order to determine the variables that differ according to the bank type, t test was performed. Return on asset and return on equity variables is used as dependent variables in the analysis. While the bank type, asset size, credit risk, liquidity risk, financial risk and operational efficiency are used as independent internal variables, macroeconomic indicators representing by inflation and GNP growth rate are used as independent external variables. As a result of the analysis; it is concluded that to made a classification as Islamic banks and conventional banks does not have an impact on profitability. This result reached in the study; (Samad & Hasan, 1999), (Samad 2004) and (Ansari & Rehman, 2012) are similar to the results of studies conducted abroad. It also resembles the results of the studies conducted by (Sakarya & Kaya, 2013) and (Doğan, 2013) in Turkey. Even if the working principles and philosophy are different, the similarity of the functioning structure of the two bank types and their operation in the same competitive environment and subject to the same legal regulations can be considered as one of the main reasons of this result. Turkish Studies - Economy, 15(2)
19. Does Bank Type Affect the Profitability ? An Econometric Research on Conventional… 745 According to the results of the t-test applied to examine whether the effects of the variables differ according to the bank type; It was determined that the variables of activity efficiency, bank size, liquidity risk and financial risk is significantly lower in participation banks than conventional banks. The fact that the operating expenses / net profit share revenues ratio, which is used to measure operational efficiency, is significantly lower in participation banks than in conventional banks shows that participation banks are better in the input-output relationship. The relatively low bank size indicates that the asset size in participation banks has not yet reached the desired level. The concept of liquidity refers to the power of banks to meet all kinds of cash needs. In the study, it is determined that the liquidity risk is significantly lower in participation banks than conventional banks. it means that borrowing items of participation banks other than deposits, for example, borrowings, commercial debts and debt to money markets, etc. are proportionally lower. In other words, the capacity or willingness to access alternative fund resources other than deposits is lower in participation banks. Determining that the ratio of equity / total assets calculated for the purpose of determining the level of financial risk is significantly lower in participation banks than conventional banks shows that participation banks are more sensitive to economic fluctuations and need to strengthen their capital structures. Developing unique and modern products in accordance with Islamic finance understanding will enable to reveal their differences within the sector. In addition, Islamic banks are expected to increase permanently their shares in the sector by paying attention to Islamic financial literacy in order to ensure the public understanding of the financial products and services that they use. So, they will be able to compete more successfully on the international platforms. References Alpar, R. (2013). Applied multivariate statistical methods. Detay Publishing. Ansari, S., & Rehman, A. (2011). Financial performance of Islamic and conventional banks in Pakistan: A comparative study. International Conference on Islamic Economic and Finance. Qatar:Doha: Qatar Faculty of Islamic Studies. Ata, H. A., & Buğan, M. F. (2016). Factors affecting the efficiency of Islamic and conventional banks in Turkey. Journal of Islamic Economics, Banking and Finance, 12(1), 90-97. Breusch, T. S., & Pagan, A. R. (1979). A simple test for heteroscedasticity and random coefficient variation. Econometrica, 47(5), 1287-1294. Cook, R. D., & Weisberg, S. (1983). Diagnostics for heteroscedasticity in regression. Biometrika, 70(1), 1-10. Doğan, M. (2013). Comparision of performances between participation and conventional banks: Evidence from Turkey. Journal of Accounting and Finance, 58, 175-188. Drukker, D. M. (2003). Testing for serial correlation in linear panel-data models. The Stata Journal, 3(2), 168-177. Er, B., & Uysal, M. (2012). Comparative effectiveness analysis of commercial banks and participation bank in Turkey: 2005-2010 term review. Atatürk University Journal of Economics and Administrative Sciences, 26(3-4), 365-387. www.turkishstudies.net/economy
20. 746 Suna Akten Çürük - Ümran Münire Kahraman - Mine Işık Fayed, M. E. (2013). Comparative performance study of conventional and Islamic banking in Egypt. Journal of Applied Finance and Banking, 3(2), 1-14. Hoechle, D. (2007). Robust standard errors for panel regressions with cross-sectional dependence. The Stata Journal, 7(3), 281-312. Im, K. S., Pesaran, M. H., & Shin, Y. (2003). Testing for unit roots in heterogeneous panels. Journal of Econometrics, 115, 53-74. Iqbal, M. (2001). Islamic and conventional banking in the nineties: A comparative study. Islamic Economic Studies, 8(2), 1-28. Jaffar, M., & Manarvi, I. (2011). Performance comparison of Islamic and conventional banks in Pakistan. Global Journal of Management and Business Research, 11(1), 59-66. Loghod, H. A. (2005). Do Islamic banks perform better than conventional banks? Evidence from Gulf Cooperation Council Countries. API Working Paper Series, 1011. Parlakkaya, R., & Akten Çürük, S. (2011). Using financial ratios to distinguish between participation and conventional banks: a case study of Turkey. Journal of Ege Academic Review, 11(3), 397-405. Rashwan, M. H. (2012). How did listed Islamic and traditional banks performed: pre and post the 2008 financial crisis. Journal of Apllied Finance & Banking, 2(2), 149-175. Pesaran, M. H., Ullah, A., & Yamagata, T. (2008). A bias-adjusted LM test of error cross-section independence. Econometrics Journal, 11, 105-127. Phillips , P. C.B., & Perron, P. (1988). Testing for a unit root in time series regression. Biometrika, 75(2), 335-346. Safiullah, M. (2010). Superiority of conventional banks&Islamic banks of Bangladesh: A comparative study. International Journal of Economics and Finance, 2(3), 199-207. Sakarya, B., & Kaya, Y. (2013). Performance differentiation between participation (Islamic) banks and deposit banks in Turkey. EY International Congress on Economics I, 24-25. Samad, A. (2004). Performance of ınterest free Islamic banks vis-à-vis ınterest-based conventional banks of Bahrain. IIUM Journal of Economics and Management, 12(2), 1-15. Samad, A., & Hasan, M. K. (1999). The performance of Malaysian Islamic bank during 1984-1997: An exploratory study. International Journal of Islamic Financial Services, 1(3). Siraj, K. K., & Pillai, P. S. (2012). Comparative study on performance of Islamic banks and conventional banks in GCC Region. Journal of Applied Finance And Banking, 2(3), 123161. Şamiloğlu, F., Tükenmez, M. N., & Bağcı, H. (2013). Comparison of the profitability performance of commercial banks and participation banks with the TOPSİS method. 17th Finance Symposium,. TBB. (2019). The Banks Association of Turkey. 10/1/2019. https://www.tbb.org.tr/en/banks-andbanking-sector-information/statistics-and-data-query/statistical-reports/20. TKBB. (2019). Participation Banks https://www.tkbb.org.tr/statistical-reports. Association of Turkey. 10/1/2019. Topaloğlu, E. E. (2017). Determination of factors affecting financial fragility in banks using panel data analysis. Eskişehir Osmangazi University Journal of Economics and Administrative Sciences, 13(1), 15-38. Turkish Studies - Economy, 15(2)
21. Does Bank Type Affect the Profitability ? An Econometric Research on Conventional… 747 Toraman, C., Ata, H. A., & Buğan, M. F. (2015). Comparative performance analysis of conventional and participation banks. C.Ü. Journal of Economics and Administrative Sciences, 16(2), 301310. TUIK. (2019). Turkish Statistical http://www.tuik.gov.tr/UstMenu.do?metod=kategorist. Institute. 10/18/2019. Tuncay, M. (2018). Comparison of financial performance of participatıon banks and deposit banks. Al-Farabi International Journal on Social Sciences, 2(1), 160-176. Usman, A., & Khan, M. (2012). Evaluating the financial performance of Islamic and conventional banks of Pakistan: A comparative analysis. International Journal of Business and Social Science, 3(7), 253-257. Yerdelen Tatoğlu, F. (2018). Panel Zaman Serileri Analizi. Beta Publishing. Yurttadur, M., & Demirbaş, H. (2017). Comparison of financial performance of participation banks and private capital conventionel banks in Turkey. IGU Journal of Social Sciences, 4(2), 89117. www.turkishstudies.net/economy