Determinants of non-performing loans in banking sector in small developing island states: A study of Fiji

Purpose: The banking sector stability depends in large part on the size of non-performing loans (NPLs). Hence, the factors which explain the problem loans are very useful information for banks. Notably, studies in this regard with respect to the small developing countries’ banking sector have received less attention. Therefore, this study aims to examine the determinants of NPLs with a case of Fiji’s banking sector, over the period 2000-2013. Design/methodology/approach: The balanced sample consists of the entire banking sector (five commercial banks and two non-bank financial institutions). First, the authors estimate a base model which comprise bank-specific indicators that are related to bank management and then they extend the estimations to include macroeconomic/structural factors such as economic growth, inflation, changes of the real effective exchange rate, unemployment, remittances, political instability and external events like the global financial crisis. The estimations are done using pooled OLS, the random effects and the fixed effects regression methods. Findings: The results show that the following indicators have negative association with NPL and are statistically significant with the conventional levels: return on equity, capital adequacy requirement, market share based on assets, unemployment and time. On the other hand, the net interest margin has a positive and statistically significant association with NPL. Research limitations/implications: Subsequently, the stability of the banking sector in small developing countries such as Fiji is largely dependent on banks’ profitability, solvency, size in terms of market share and the presence of a learning curve and keeping a close tab on the interest rate spread between loans and deposits. Practical implications: The paper highlights the specific factors determining NPL in small developing economy of Fiji. Originality/value: This study is the first to examine specific factors determining NPLs with respe