Monetary policy stance and foreign currency lending: evidence from a persistently dollarized emerging market

Abstract

This study explores how central bank policy changes influence Turkish banks’ foreign currency lending in one of the most dollarized emerging markets. Using quarterly data from 2002Q4 to 2024Q3 and applying a Bayesian Additive Regression Trees model within a machine learning framework, we uncover that higher policy rates amplify foreign currency lending—especially among larger banks—thereby weakening monetary transmission. The analysis also highlights the nonlinear and asymmetric effects of deposit dollarization, bank size, and macroeconomic conditions. These findings offer new insights into the structural constraints of monetary policy under persistent dollarization, with direct implications for policy design in similar economies.