Decoupling and recoupling in the crude oil price benchmarks: An investigation of similarity patterns

The aim of this paper is to investigate the decoupling and recoupling of WTI and Brent prices also with respect to the debate on the regionalisation-globalisation of the two oil markets.To this purpose, we employ the Dynamic Time Warping (DTW) algorithm to identify decoupling events between the two crude oil price series. DTW has been employed for classification and clustering aims in many fields, but in this paper we make a slightly different application of DTW with respect to those provided by the literature, demonstrating how DTW can be employed also to investigate the decoupling between the two oil benchmarks. Our analysis reveals that the two oil benchmarks decouple and recouple according to WTI local market conditions. Therefore, we found evidence that WTI-Brent market is not fully integrated at all times. We also propose two DTW-based indexes: Relative-Alignment Index (RAI) and Warping Index. The first confirms that the greatest decoupling between WTI and Brent occurs because of WTI local market conditions and is useful in highlighting the main decoupling between our crude oil series over time, while the second provides information on the time window of crude oil price decoupling. Lastly, we provide some policy implications based on our results. •We investigate the decoupling-recoupling of WTI/Brent prices (regionalisation-globalisation of both oil markets).•We employ the Dynamic Time Warping algorithm to identify decoupling events between the two crude oil price series.•The two oil benchmarks decouple and recouple according to WTI local market conditions.•We propose two DTW-based indexes: Relative-Alignment Index (RAI) and Warping Index.•RAI confirms the role of WTI local market conditions. WI gives information on the time window of crude oil price decoupling.