Housing prices and mortgage interest ratesToda-Yamamoto causality test

Abstract

This study analyzed the causality between house prices and mortgage interest rate. Study term is 2010:01-2015:04 because of the fact that the house price indexes were calculated since 2010. We applied the Toda-Yamamoto (1995) causality test which eludes the shortfalls associated with the standard Granger (1969) causality test. The study also estimated the impulse response functions and variance decompositions. The MWALD results revealed one way causality between mortgage interest rate and house price index, where the former Granger causes the latter. The reverse is not correct. Same results are valid for new house price index. According to the impulse response function results, shocks to the mortgage interest rate have a lagged and significantly negative impact on both the house price index and new house price index. Variance decomposition results show that both house price index and new house price index have been explained by their own lag values and the values of mortgage interest rate. These results contain some conclusions about Turkish housing markets for both academicians and practitioners.