Predicting the future financial performance of Islamic banks: a sentiment analysis approach

Purpose - This study aims to predict the financial performance of Islamic banks with sentiments of management from the textual information in annual reports. Design/methodology/approach - The study uses data from 33 Islamic banks in six Islamic countries from 2006 to 2020. The authors estimate the model using the system GMM because it helps dealing with endogeneity problem, which are inherent in panel data. Findings - The findings of the study reveal that there is a strong relationship between the sentiment expressed by management in annual reports and the current (future) financial performance of Islamic banks. The higher the positive sentiments of management, the better financial performance. In addition, the study also suggests that negative sentiments using term frequency-inverse document frequency is linked to a decrease in banks’ financial performance. Research limitations/implications - The study does not present the Islamic view on sentiment analysis in the context of Islamic scriptures due to the unavailability of a relevant dictionary. Practical implications - The findings of the study suggest that developing accurate models with the help of textual information for performance prediction of Islamic banks help shareholders, regulators and policymakers avoid devastating events. Using textual information may also help reduce the information asymmetry between the management and shareholders, which may lead to more efficient bank supervision. The study can also help investors evaluate their prospective investments in the Islamic bank. Originality/value - To the best of the authors’ knowledge, this study is the first of its kind that uses management sentiments for performance prediction of the Islamic banking sector. It may add a valuable contribution to the existing literature.