Interest-free versus Conventional banks- A Comparative Study using Linear and Nonlinear Panel Regression: Empirical Evidence from Turky and 6 MENA countries

This paper contributes to the empirical literature on interest-free finance by investigating the feature of interest-free and conventional banks in Turky and 6 MENA countries over the period 2005–2014. To distinguish between interest-free and conventional banks [in terms of Profitability, Liquidity, Credit and Insolvency risk, and Stability], we use two-sided t-test, linear regression model, Non linear Panel model (Random Logit and Pooled Probit), and Discriminant function analysis. Using a sample of 115 banks (80 conventional and 35 interest-free banks), univariate results based on t-test show that interest-free banks (IB) are, on average, less profitable, more liquid, less stable, and have higher credit risk but are more solvent, than their conventional peers (CB). We find also that the difference between the 2 banking types was significant pre and post the GFC. IB are more profitable Pre GFC and more solvent post GFC. Results from linear regression models show that the two types of banks may be differentiated in terms of, bank characteristic, Size, Cross countries, and Maket Share. Small IB are more profitable, more capitalized, and more stable than Small CB (with or without islamic window). From the Pooled Probit model (Random Logit) results, banks which have more liquidity, which are better capitalized, more solvent, and which are less stable (less stable), are more likely to be IB. We find also that there is no difference between pre and post the GFC. From Discriminant function analysis, AGE was the strongest predictor in discriminating the two types of banks while Z-score was the next in importance as a predictor.