Covered Interest Parity Deviations Between India And The US

Abstract

Prior studies have tested Covered Interest Parity (CIP) between India and the United States and found substantial deviations. The main objective of the current study is to econometrically model and explain deviations from CIP. Further, the study contributes to the literature by proposing an approach to testing CIP after allowing for country risk. A preliminary analysis suggests that there are two types of shocks that impact the CIP deviation, also referred to as the Covered Interest Differential (CID): permanent shocks and temporary shocks. The permanent shocks may be interpreted as reflecting a change in the country risk premium and the temporary shocks as reflecting transient effects and disequilibrium. The paper uses a bivariate Vector Autoregression (VAR) approach to model the joint dynamics of the CID and the forward premium, and applies the methodology of Blanchard and Quah (1989) to separate the impact of the two types of shocks. Impulse-Response analysis shows that a one standard deviation permanent shock has an immediate, substantial impact on the CID. However, forecast error variance decomposition reveals that less than 30% of the variability in the CID is caused by such permanent shocks. Further, permanent shocks account for less than 5% of the forecast error variance of the forward premium, which suggests that covered interest arbitrage activity has limited influence on the forward premium. Temporary shocks appear to be related to transient volatility in the forward premium, and such shocks initially affect both the forward premium and the CID to approximately the same extent. The manner in which the CID responds to a temporary shock suggests considerable impediments to arbitrage. However, the fact that the CID recovers at a slightly faster rate than the forward premium, especially in the initial periods, suggests that capital restrictions are not completely binding.