Purpose
The purpose of this paper is to show that in some industries the linear model may not reveal any significance link between exchange rate volatility and trade flows but once nonlinear adjustment of exchange rate volatility is introduced, the nonlinear model reveals significant link.
Design/methodology/approach
This paper uses the linear ARDL approach of Pesaran et al. (2001) and the nonlinear ARDL approach of Shin et al. (2014) to assess asymmetric effects of exchange rate volatility on trade flows between Germany and Turkey.
Findings
This paper consider the experiences of 75 2-digit industries that trade between Turkey and Germany. When the study assumed the effects of volatility to be symmetric, the study found short-run effects in 31 (30) Turkish (German) exporting industries that lasted into the long run in only 10 (13) Turkish (German) exporting industries. However, when the study assumed asymmetric effects and relied upon a nonlinear model, the study found short-run asymmetric effects of volatility on exports of 55 (56) Turkish (German) industries. Short-run asymmetric effects lasted into long-run asymmetric effects in 10 (25) Turkish (German) exporting industries. All in all, we found that almost 25% of trade is hurt by exchange rate volatility.
Originality/value
This is the first paper that assesses the possibility of asymmetric effects of exchange rate volatility on German–Turkish commodity trade.
This paper investigates possible asymmetric influences of the exchange rate on cross-border freight flows between the U.S.A. and Canada. Linear and nonlinear autoregressive distributed lag models are used to test for the existence of long-run asymmetric effects of 1) currency appreciation and depreciation and 2) exchange rate volatility changes on trade flows by truck, rail, air, vessel, and pipeline. This paper provides evidence that both currency value and exchange rate volatility affect the U.S.–Canada freight flows in an asymmetric manner. The long-run results of the nonlinear models show that exchange rate is found to be significantly associated with the bilateral trade flows between the U.S.A. and Canada. Exchange rate volatility tends to be significantly associated with trade flows in the nonlinear models, while its effects are insignificant in most cases in the linear models. These findings suggest that the conventional linear specification may mislead the asymmetric effects of exchange rate uncertainty on cross-border freight flows. It is also found that exchange rate sensitivities of U.S.–Canada trade flows by transport mode can differ significantly from those of aggregate trade flows. The information derived from disaggregate trade data can be useful for traders and shippers to develop a long-term strategic plan for infrastructure investment and service expansion.
China is the largest trading partner of Pakistan. Therefore, it is very important to consider the trade flows between Pakistan and China and their response to rupee–yuan volatility. Previous research assumed that response of trade flows to measure of volatility is symmetric. In this study, our basic objective is to check whether the trade flows respond to volatility in a symmetric or asymmetric manner. Annual data over the period 1980–2018 for 14 Pakistani industries exporting to and 34 industries importing from China are analyzed. We find short-run asymmetric effects of exchange rate volatility in almost all industries that last into long-run asymmetric effects in 40–50 per cent of industries. Non-linear models yielded more significant effects of volatility than the traditional linear models.
Keeping in view the idea of the third-country effect by Cushman, the analysis attempts to capture the asymmetric impact of third-country exchange rate volatility on Pakistan–China commodity trade. The empirical analysis is based on the annual data for 14 industries that export from Pakistan to China and 34 industries that import to Pakistan from China. The findings of the study confirm that nonlinear models generate more significant results both in the short and long runs. Moreover, the empirical findings suggest that the asymmetric assumption alone is not enough, and instead, we should use it along with the third economy effect.