Interlinkages among Exchange Rate, Interest Rate, Consumer Price Index, and Output Volatilities

Rashid Rauf and Abdul Rashid

Abstract

In this paper, we investigated the interlinkages between real exchange rate volatility (VRER), consumer price volatility (VCPI), industrial output volatility (VLMI), and interest rate volatility (VINT) using monthly data covering the period January 1988-December 2017. We applied a multivariate model of Granger causality and found bidirectional causality between VRER and VLMI. Similarly, we found bidirectional causality between VRER and VINT. Whereas, unidirectional Granger causality running from VCPI to VRER, VLMI to VCPI, and VINT to VLMI. The results of variance decomposition show that the VRER error forecast is mainly attributed to its own shock. However, after 18 months it reaches to 94.6% while the remaining 5.4% error forecast is explained by VLMI, VCPI, and VINT shocks. In a similar manner, at a time horizon of 1 month, 99.5% of the VINT error forecast is associated with its own shock but after 18 months it declines to 89.6%. As expected, the VLMI error forecast is mainly attributed to its own shock and it declines to 97.2% after 18 months. Contrary to this, at a time horizon of 1 month, 97.6% of the VCPI error forecast is attributed to its own shock, whereas after 18 months, it rises to 98.8%, remaining 1.2% is due to VRER, VLMI, and VINT shocks. The impact of VRER on VRER, VCPI, and VINT is positive but it fades away with a short passage of time. Likewise, the effect of VCPI on VRER and VCPI is positive and convergent. In contrast, the response of VRER and VINT is positive and persistent. The response of all the macroeconomic series, except VLMI to VINT, is convergent.