Bringing 'Money' Back in Monetary Models of Exchange Rate

Abstract

Abstract Money overtime has been deemphasized from most of the open-economy macroeconomic models of exchange rate. The primary reason being the growing instability of money demand due to the rapid pace of the financial innovation and the velocity slowdown post 1980s. As money began to get de-emphasized, interest rate alone became the monetary policy instrument. However the identification of the monetary policy requires the knowledge of both money supply and the money demand. In other words, interaction between the Central Bank’s reaction to economic condition and the private sector’s response to the policy action is crucial to the correct monetary policy identification, better prediction of the equilibrium quantity of the money and future rate of interest. Therefore attempts have been made to incorporate monetary aggregates back in the monetary models of the exchange rate determination. Additionally, a superior monetary measure, the aggregation-theoretic Divisia monetary aggregate has been introduced. Divisia money provides an index of ‘monetary services’ which captures the traditional transaction motive for holding money i.e. money demand. Exchange Rate Overshooting in Small Open Economies: A Reassessment in a Monetary Framework- There has been a remarkable drop in the commodity prices and currency weakening for all the resource reliant economies since mid-2014. Rethinking on the central bank’s monetary policy and reassessing the response of the exchange rate to policies is an active area of research. In the words of Kenneth Rogoff, “If one is in a pinch and needs a quick response to a question about how monetary policy might affect the exchange rate, most of us will still want to check any answer against Dornbusch’s model.”1 Surprisingly studies on monetary policies have found exchange rate effects that are mostly inconsistent with Dornbusch’s overshooting hypothesis. Bjornland (2009) finds evidence of exchange rate overshooting by using identification restrictions that acknowledge simultaneity between monetary policy and exchange rate. However, the correct way of identifying monetary policy requires meticulously capturing the central bank’s reaction to economic condition and at the same time private sector’s response to policy action. This calls for introduction of ‘monetary’ aggregates back in the monetary model of exchange rate determination. Motivated by the Bjornland’s result, this paper rediscovers the validity of Dornbusch Overshooting hypothesis for Australia, Canada, New Zealand and Sweden. More specifically, a contractionary monetary policy shock leads to exchange rate overshooting as predicted by Dornbusch. The exchange rate appreciates significantly on impact to a monetary policy shock as shown by the impulse response functions and thereafter depreciates. Also the variance decomposition results show that money demand and money supply shocks explain a significant portion of exchange rate fluctuations vis-a-vis Bjornland’s original model. An SVAR Approach to Evaluation of Monetary Policy in India: Solution to the Exchange Rate Puzzles in an Open Economy- Following the exchange-rate paper by Kim and Roubini (2000), the questions on monetary policy, exchange rate delayed overshooting, the inflationary puzzle, and the weak monetary transmission mechanism are revisited; but this is done so for the open Indian economy. A superior monetary measure, the aggregation-theoretic Divisia monetary aggregate is incorporated in the model. The paper confirms the efficacy of the Kim and Roubini (2000) contemporaneous restriction, customized for the Indian economy, especially when compared with recursive structure, which is damaged by the price puzzle and the exchange rate puzzle. The importance of incorporating correctly measured money into the exchange rate model is illustrated, when compared among the models with no-money, simple-sum monetary measures, and Divisia monetary measures. The results are confirmed in terms of impulse response, variance decomposition analysis, and out-of-sample forecasting. In addition, a flip-flop variance decomposition analysis is done that suggests two important phenomena in the Indian economy: (i) the existence of a weak link between the nominal-policy variable and real-economic activity, and (ii) the use of inflation-targeting as a primary goal of the Indian monetary authority. These two main results are robust, holding across different time period, dissimilar monetary aggregates, and diverse exogenous model designs. Divisia monetary model of exchange rate determination: A Multi country Analysis- There was a breakdown in the stability of money demand after the mid-1970s, an era characterized by financial liberalization and financial innovations. Therefore it simply became convenient to do away with the monetary aggregates and consider interest rate as the sole monetary policy variable. However it will be misleading to measure the impact of monetary policy and thereby rightly track the monetary policy transmission mechanism with the interest rates alone (when the rates are already stuck near zero) especially post 2008-crisis. Also any model that explicitly capture the role of ‘money’ can aid the central bank to track the money demand behavior of the private agent. This is crucial to monetary policy identification and measuring the impact of monetary policy on macroeconomic variables. Divisia money provides an index of ‘monetary services’ which captures the traditional transaction motive for holding money i.e. money demand. A modified Wald test criterion suggested by Toda-Yamamoto is taken up in evaluating the information content of the Divisia monetary aggregate along with the alternative monetary policy indicators. The findings challenge the traditional view that short-term/ immediate rate of interest have better predictive power on the exchange rate than the alternate monetary indicators. Subsequently Divisia monetary aggregate has been used in the monetary models of exchange rate and compared with model setup containing their simple-sum counterparts. Using both money supply and money demand, the role of ‘money’ has been evaluated in the structural vector autoregression setup for India, Poland and United Kingdom. The results of the study show that models with money especially Divisia money (i) help correctly identify shock to monetary policy (ii) shows more significant response of exchange rate to policy shock (obtained by Random Walk Metropolis Hastings method of Bayesian Monte Carlo integration) (iii) remove some of persistent puzzles like price puzzle, exchange rate puzzle and forward discount bias puzzle (iv) facilitate policy variables explain more of exchange rate fluctuations (v) generate better out-of-sample exchange rate forecast values in terms of RMSE and Theil U (obtained by estimating the model using Kalman filter) and (vi) generate better out-of-sample forecast graphs for exchange rates (obtained through Gibbs Sampling on a Bayesian VAR with a “Minnesota” prior)