PurposeThis paper aims to shed light on the risk structure in the presence of Islamic banking. The author concentrates on the relationship between Islamic banking and conventional banking in Turkey. Islamic banking and conventional banking are considered to be different kinds of sources for funding. Returns in the conventional banking are expected to be heavily influenced by the interest rate in the money market. However, Islamic banking returns are interest-free so that interest rate changes are not expected to affect the deposit returns in Islamic banks. Interest rates in the economy are a proxy to highlight the general risk level of the economy. By looking at the causal relationship between the deposit returns of both Islamic banks and conventional banks, it is possible to address the different types of banking in the general risk structure of the economy. This is one of the first studies to address the mentioned difference in banking sector in Turkish economy.Design/methodology/approachThis paper tries to identify the direction of causality between Islamic and conventional banking term deposit rates by means of Granger Causality. Also, Granger Causality test results will guide to explore the Islamic and conventional banking deposit return linkages. The author has extended the study with vector autoregressive analysis to understand the correlation structure between conventional deposit rates and the profit–loss sharing ratio of Islamic Banks. The author has also extended this study with impulse response functions to see whether the shocks hitting into the conventional banking affect Islamic banking and vice versa.FindingsThe results suggest that there is no significant clear relationship between both banking sectors. This result can be interpreted, as Islamic banks do not adjust their profit–loss sharing (PLS) ratios pegged to the interest rate offered by conventional banks. Also, conventional banks determine their interest rate without any connection to the Islamic banking PLS ratios. Overall results of this study contradict the findings of studies which conclude that Islamic banking might not be different from the conventional banking. It is reported that inferences from pair-wise Granger causality alone might be spurious, as the analysis based on non-stationary series can be a consequence of time functional characteristics of the time series.Social implicationsThe results can be taken as counter evidence to the hypothesis “Islamic banks determine their PLS ratios based on the interest rates offered by conventional banks”. This address that the Islamic banks may offer alternative financing methodology which has different procedure. Hence, Islamic finance can be taken as an alternative method with its asset-based healthier structure.Originality/valueThis is one of the first studies to address the Islamic versus interest-based banking difference in banking sector in Turkish economy. This paper tries to identify the direction of causality between Islamic and conventional banking term deposit rates by means of Granger causality.