Long Term Relationship between Crude Oil Price and Exchange Rate in the context of India and Emerging Countries

Avik Ghosh^*
*Correspondence: Avik Ghosh, Department of Management Sciences, The Institute of Cost and Management Accountants of India, Karnataka, India, Email:

Keywords

Crude oil • Exchange rate • Variance decomposition • Vector auto regressive • Shock analysis

Introduction

The economic health and well-being of any country, be it a developed or a developing one, depends on its macroeconomic state which is more or less identified with some well-established indicators namely GDP growth, fiscal deficit, inflation, exchange rate, gross capital formation, employment scenario etc. However, these indicators are interdependent in nature where a slightest variation in one impacts the movement of others. The impossible trinity or Mundell-Fleming Trilemma of free flow of capital, fixed exchange rates and independent monetary policy is a reflection of the phenomenon. While ascertaining such dependence of mutually exclusive variables, it has been assessed that some of them maintain strong causality (both unidirectional and bidirectional) with external variables.

India’s current account deficit has increased from 1.8% of GDP FY 2018 to a whopping 2.1% FY 2019 resulting in a net outflow of $57.2 Billion. This not only reflects the health of India’s import-dependence but also is a matter of concern when the expected currency depreciation can be foreseen. The total import of approximately $450 Billion and a considerable 25% contribution from the import of crude oil depicts the extraordinary crude oil centric import dependency of Indian economy (Table 1). The related outflow of foreign exchange weakens the fiscal structure of economic framework by reducing the demand of home currency compared to foreign currency. Hence the variation in crude oil price impacts the exchange rate to a great extent. Any shock on the crude oil price also results in a subsequent change in exchange rate with an interdependent lag structure.

We cannot say in the absolute terms that this is the result of impact of oil prices in global market and import dependency of emerging market economies on crude oil. Past literature review gives us evidence on the short term dependency of exchange rate on oil prices shocks.

In this research, we are going to examine the impact of the oil prices on exchange rate taking into fact the inflation effect by bring out into consideration the real exchange rate and WTI oil prices deflated by CPI inflation of US. So that we can measure the long term and short term implications of oil prices on exchange rates in real terms.

Literature Review

Oil exporting country experiences exchange rate appreciation in response to soaring oil prices and vice versa. It had also suggested that oil importing countries may experience depreciation in exchange rate when oil price rises and vice versa. The work of Hamilton in 1983 had shown the connection between the changes in oil prices and US business cycle. The oil being a homogeneous and globally traded commodity priced in US dollar had shown that depreciation in dollar increases the purchasing power of foreign currency which in turn increases the price of crude oil through the demand effect [1].

The positive shocks to oil prices depress the emerging market stock price as well as US dollar exchange rates in the short run. They also examined that positive oil production shock lowers oil prices and a positive shock to real economic activity increases oil prices which is evident by the empirical theory of demand and supply [2].

The long-run relationship between real oil prices and real exchange rates by using a monthly panel of G7 countries from 1972 to 2005. They had shown that real oil prices may have been the dominant source of real exchange rate movements and there is a link between real oil prices and real exchange rates. They had also examined whether real oil prices can be used to predict real exchange rates. In the same year 2007. Observed the impact of oil on real effective exchange rate of dollar between 1974 to 2004 and revealed that approximate 10% rise in oil prices result in nearly 4.3% appreciation of the dollar in the long run for an oil exporting country [3].

The oil price and exchange rate relationship for India using daily data for the time span from July 2^nd , 2007 to November 28^th , 2008 using GARCH and EGARCH approach and have found that an increase in the oil price leads to the depreciation of Indian currency compared to US dollar. The study also establishes that positive and negative oil price shocks have similar effects in terms of magnitude on exchange rate volatility.

Continuous wavelet approach and deployed asymmetric, multihorizon granger-causality tests between the return series of the oil price and the India-US exchange rate, over the time-span 1980-2016. They have highlighted the co-movements of both the variables in the post-reform period, especially in 2-4 years band and in the short run causality runs from oil to exchange rate. They had also found that Since a lot of research has been done on the relationship between oil prices and exchange rates, this paper consists of an analysis from January 1994 to April 2019 between the real exchange rate return and real crude oil price return analyzing the short term and long term impact of both the variables on each other.

Discussion

Initial theoretical framework

As the objective of this study is to find out the causality and relationship between the two variables of crude oil price and exchange rate, a multivariate model called Vector Auto Regressive model (VAR) has been considered as the basis of analysis. Monthly data has been analyzed from January 1994 to April 2019 and a total of 304 data points have been taken into consideration. The log return of the month wise data series has been considered for this purpose to normalize the dataset and to reduce the variances. Real broad effective exchange rate for India has been considered as exchange rate data whereas CPI adjusted global price of WTI crude has been used as crude oil price [5].

Once the comparable data set is prepared, the analysis has been performed estimating the parameters with vector auto regressive model. Then the VAR model lag order selection criterion has been implemented to assess the significance of the lag orders. Lag exclusion Wald test followed to reiterate the dependency need of the parameters. Granger causality/block exogeneity Wald test was executed then to establish the bidirectional causality.

As stated in the methodology, the data analysis started with the estimated equation for Vector Auto Regression (VAR) model where both the log returns of exchange rate and crude oil price data was considered as endogenous variables. While assessing the outcome of VAR model, it is found from the t-statistics that for both the variables, some of the lags are significant in defining the nature and movement of the variables. For establishing the veracity of the actual lag, we then executed lag exclusion Wald test with an aim to assess the correct lag order [6].

This has been performed for two lags where the hypothesis is:-

• H0: The lag order can’t be excluded

• H1: The lag order can be excluded and the next lag order can be tested.

The joint p-statistics of lag order 1 is very much significant resulting in rejection of Null hypothesis and the need to test second order lag. The joint p-statistics for second order lag is not significant at 5% level and establishes the null hypothesis. Hence the dilemma from the lag length criteria to select the lag order is statistically resolved with the lag exclusion test establishing the second order lag.

The granger causality block exogeneity Wald test was performed next which is proved to be instrumental for the outcome of the paper. The chi-square values and the probabilities of bidirectional testing for both exchange rate and oil price return can’t reject the null hypothesis of no granger causality. Both the test statistics clarify that although the two variables are having some significance with their lag orders, they are not granger causal (Table 2) [7].

The error variance of both the variables is self-dependent and can’t be ascertained by exogenous shocks to the other variables. With these we establish the fact of block exogeneity and noncausality of both these variables in long term although they have significant lag relationship.

The above analysis examines the monthly data spanning from January 1994 to April 2019 for India. Real exchange rates (effective) are found from the RBI website database. Crude oil prices are obtained from data available at FRED and it is further deflated by US consumer price index to do away with the effect of Inflation. Further the return of real effective exchange rate and real oil prices are evaluated and the natural logarithm is used to normalize the effect of residuals errors.

Further, we can examine the other variables which can impact the exchange rate i.e. a representative variable to indicate the competitiveness of a county. We may use variables like expenditure in research and development, capital investment as a percentage of GDP. These micro variables impact the exchange rate of emerging market in the long run. Investment in research and capital expenditure may result in raising the competitiveness of country export and will be able to drive the exchange rate in long terms.

In terms of limitations, exchange rate could be influenced by global shocks like economic recession, emerging market crisis, and bruit events. These subjective variables were difficult to incorporate in our research.

Conclusion

Our research started with a defined goal to have a thorough crosscountry study of the relationship between exchange rate and crude oil price. Various related research papers from countries with different economic profile have been studied and all aspects pertaining to the causality were identified. Some of the countries depicted auto regression; few have shown short term causality whereas many of them depicted either non-causality or minor interdependency. The similar analysis was performed for India where the outcome initially hinted a relationship of the endogenous variables but subsequently proved to be non-causal. This outcome emancipates the very fact about the Indian currency that it is capable of absorbing crude oil price shock whereas the future error variance of the exchange rate doesn’t depend on crude oil price. This shows that various macroeconomic factors ranging from fiscal and monetary prudence contribute considerably in the exchange rate variation in long term. This non-causality in an emerging market like India is an eye-opener as the forecasting of exchange rate, being non-causal with crude oil price, can be performed with its own past value or with autoregressive modeling of other suitable parameters.

References

1. Aboody, Golub, Stephen S. “Oil Prices and Exchange Rates”. Econ J 93 (1993): 576-593.
2. Blomberg, S Brock and Harris Ethan S. "The Commodity-Consumer Price Connection: Fact or Fable?." Econ Policy Rev 1 (1995): 21-38.
3. Amano, Robert A., and Simon Van Norden. "Oil Prices and the Rise and Fall of the US Real Exchange Rate." J Int Money Fin 17 (1998): 299-316.
4. Chaudhuri, Kausik, and Daniel Betty C. "Long-Run Equilibrium Real Exchange Rates and Oil Prices." Econ Lett 58 (1998): 231-238.
5. Sadorsky, Perry. "The Empirical Relationship Between Energy Futures Prices and Exchange Rates." Ene Econ 22 (2000): 253-266.
6. Udell, Gregory F. "What's in a Relationship? The Case of Commercial Lending." Business Horizon 51 (2008): 93-103.
7. Berger, Philip G., Minnis Michael and Sutherland Andrew. "Commercial Lending Concentration and Bank Expertise: Evidence from Borrower Financial Statements." J Account Econ 64 (2017): 253-277.