The Determinants of Nonperforming Loans: The Case of Turkey

Abstract

The aim of this paper is to investigate the macroeconomic and sector-specific factors likely to affect nonperforming loans (NPLs) for the Turkish banking sector. The study covers a quarterly time series span of 2006–2015 and uses the time series econometric methods of the Johansen cointegration test, vector error correction model (VECM), and the Granger causality test. In the analysis, we use the industrial production index (IPI), top 100 companies among the Borsa Istanbul Stock Exchange Index (BIST 100), and exchange rates between the Turkish lira to euro (EUR) and Turkish lira to U.S. dollar (USD) as proxies for the state of the economy and bank-specific factors; return on asset (ROA) and return on equity (ROE) are proxies for the managerial efficiency of banks in determining NPLs. The general findings support the relevance of both micro- and macroeconomic fundamentals significantly affecting Turkish NPLs. In this study we also illustrate the adverse impact of external foreign currency financing on the ability to service debt for Turkish credit market participants. The findings may prove to be helpful when designing macro- and microeconomic policies aimed at mitigating systematic risk.