ABSTRACT
The existing literature on demand for crude oil in developed and developing countries typically assumes that the effects of oil price changes are symmetric. In this paper, we use the nonlinear autogressive distributed lag (ARDL) approach of Shin et al. (2014) and test whether the effect of oil price changes on demand for imported crude oil is symmetric or asymmetric in the case of Korea. Using quarterly data over the last two decades, the results show that the oil price effects are indeed asymmetric in the long run; Korea’s imported demand for crude oil is more responsive to oil price spikes than to oil price plunges. However, the asymmetric effects are not observed in the short run.
Notes
1 It is worth mentioning that recently, Umekwe and Baek (Citation2017) provide supportive evidence of the asymmetric oil price effect in the case of U.S. shale oil production.
2 Since the F-test can be used regardless of whether a series follows I(1) or I(0) process, a unit root test is not a prerequisite for the application of the ARDL.
3 Diagnostic tests for serial correlation is also used as another criterion in selecting the maximum lag length. With a maximum of four, for example, the null hypothesis of no serial correlation can be rejected. With a maximum of two, on the other hand, the null cannot be rejected. Thus, a lag length of two quarters is chosen for further analysis.
4 This finding is consistent with the result of Kim and Baek (Citation2013).