Measuring Currency Risk Premium: The Case of Turkey

This study examines the determinants of the change in currency expectations with different maturities (1-month, 3-month, and 1-year) in the Turkish Lira (TL) versus the US dollar. The risk premium is estimated using the interest rate differential and a latent component called the missing risk premium. The empirical model is extended to break down the risk component further using other explanatory variables, such as currency swap agreements, credit default risk (CDS), foreign reserves, and the VIX (Volatility Index). This study used a state-space model to deal with unobserved variables or parameters. The empirical model is evaluated between January 2005 and March 2023 with daily and weekly frequencies. The study's findings do not support the uncovered interest parity (UIP) condition. Instead, they favour the outcome of Fama's (1984) exchange rate prediction regressions. Our findings indicate interest rates and swaps explain most of the variation in the currency risk premium in the TL. Moreover, we found a significant increase in both the level and volatility of the missing risk premium for long maturities after 2018. The coefficient of the missing risk premium and its long-lasting impact of the shock is significantly reduced when observable variables are added. Overall, this study sheds light on the complex interplay between changes in monetary policy, exchange rate, and risk premia in an emerging market context.