The Impact of Exchange Rate and Oil Price Fluctuations on Iran's Agriculture Trade Balance: Application of the J Curve Approach

Exchange rate and oil prices are the important factors for foreign trade in any country and even fluctuation in these variables will affect the economic and trade growth. The purpose of this study is to investigate the effect of exchange rate and oil price fluctuations on trade balance of Iran's agriculture sector with its 8 major trading partner over the period 1998 to 2017 and examine also the existence of the J Curve in these countries. To this end, linear and nonlinear ARDL models were utilized based on literature and tried to determine lung-run and short run effect of underling variables. Then, the results of linear and non-linear ARDL models were compared.
Materials and Methods

Since eight countries, including the United Arab Emirates (UAE), Iraq, Afghanistan, Turkey, Korea, India, Germany and China are Iran's largest trading partners during 1998-2017, we focused on these countries. In this context, the model proposed by Oskoee et al. (2011) has been used to evaluate the impact of exchange rate and oil price fluctuations on agriculture trade balance. To capture the exchange rate and oil price fluctuations, the GARCH family models were applied (including EGARCHT, GARCH, SAGARCH, and NGARCH). Time series of exchange rate and oil price fluctuations which are extracted from GARCH models, are expected to be stationary. So, according to the empirical studies, the ARDL model is an appropriate model. However, both linear and nonlinear ARDL models were estimated. To specify trade balance equation, variables including Iran's GDP, GDP of eight trading partner countries, exchange rate, Oil prices fluctuations, exchange rate fluctuations and economic sanctions have been used. We used the ADF unit root test to check stationary of the variables.

The estimated results of the GARCH family models show that the sum of the coefficients of α+β for Turkey, Iraq, India, China, Afghanistan, Germany, and Korea are 0.88, 0.94, 1, 0.92, and 0.82, respectively. As the sum of the coefficients must be between 0 and 1, the predicted fluctuations series of exchange rate are stationary and also the predicted fluctuations series is as well. After obtaining fluctuations series of exchange rate and oil price, the number of optimal lags should be determined in ARDL model. According to the FPE criterion, the optimal lag is two and according to the AIC and SBC the optimal one is three.  Since the number of observations is low, the optimal lag number was selected two and the Linear and non-linear ARDL model was estimated. The results revealed that if Iran's GDP increased by 1%, the trade balance between Iran and Turkey would  improve by 18.20% and this value for Iraq, Afghanistan, UAE, China, Germany, Korea would be 59.07, 8.40, 26.28, 91.17, 16.32, 0.16, 22.02 respectively. In the long run, if Turkey's GDP rises 1%, the trade balance between Iran and Turkey will improve 14.34%. Moreover, if GDP in Iraq, Afghanistan, UAE, Chinese, German, Korean climb by 1%, the trade balance reaches 38.31, 7.003, 10.41, 17.99, 0.39 24.6 respectively. If the exchange rate rises 1% in Iraq and Germany, the trade balance will improve roughly 26.9, 69.4 respectively. Escalating National currency in Turkey and India has reverse effects on the trade balance. In fact, as the exchange rate rises, imports from Turkey and India increased and this contradiction may be due to sanctions and economic conditions. In China and India, positive and negative fluctuation has positive and negative effects on the trade balance. Indeed, by increasing the positive exchange rate fluctuations, the trade balance would improve and with the negative exchange rate fluctuations (the exchange rate decline) the trade balance might worsen. In the nonlinear ARDL method, exchange rate fluctuations in India and China are positive and have significant effect, and it shows that there is a j-curve between Iran and these countries. Also separating exchange rate fluctuations in positive and negative groups can prove the existence of the j Curve.

According to the results, the highest value of agricultural exports is related to Iraq and the least is to Korea. The UAE has the highest imports from Iran and Iraq has the lowest one. The co-integration test reveals that the underlining variables follow and influence each other in the long run. Based on previous studies and predicted signs for coefficients of the variables in the models, the non-linear ARDL model provides better results. The finding showed that GDP of 8 countries were positive and had significant effects and Iran's GDP was negative and significant in these eight countries. In the long run, an oil price fluctuation in Turkey, Afghanistan, Germany and India has positive and significant impact. In fact, as oil prices increase, the agricultural trade balance improves. In the short run, as oil prices rise, the agricultural trade balance would decline in countries such as Turkey, Germany and India and increase in Iraq and China. By dividing the exchange rate fluctuations into positive and negative parts, we conclude that positive exchange rate fluctuations in China and India have a positive effect on the trade balance and negative fluctuations have a negative effect on the trade Balance. The current study confirmed the existence of the J curve in India and China.