THE VALIDITY OF THE NEO-FISHER HYPOTHESIS FOR THE TURKISH ECONOMY IN THE PERİOD 2004-2020

Abstract

The Fisher equation reveals that real interest rates do not change in the long run and that there is a one-to-one relationship between expected inflation and nominal interest rate. In this context, the Fisher hypothesis claims that there is a causality from inflation rate to interest rate. However, in many studies conducted after the 2008 crisis, it has been concluded that there is causality contrary to the Fisher hypothesis, especially in countries such as the USA, Japan and the EU with inflation and interest rates close to zero. In the literature, this situation is called the Fisher paradox or the Neo-Fisher hypothesis. In this study, the direction and nature of the relationship between nominal interest rates and inflation for the Turkish economy between 2004 and 2020 has been tried to be discussed in different periods. For this purpose, after investigating the relationship between interest rates and inflation with the Granger causality and VAR model, the parameters of the Taylor Equation were obtained by the Rolling Windows method and the Neo-Fisherian proposition was tested over these parameters. Accordingly, as a result of the analyzes made in the period, it was concluded that inflation was the Granger cause of interest rates, but interest rate was not the Granger cause of inflation. Similarly, when the impulse-response functions obtained as a result of the VAR analysis are taken into account, it is observed that a one-unit shock increase in interest rates lowers inflation, while a one-unit shock increase in inflation increases interest rates. Finally, according to the time-varying parameters obtained by the Rolling Windows method, the Neo-Fisher hypothesis is only valid for the period 2013M05-2014M07 in Turkey.