اثرات معاملات نیمکس بر بازارهای آتی نفت خام برنت آی پی ای : تجزیه و تحلیل دوره ای

Recent developments in the energy markets, and the surge and dip in crude oil prices over the last few years, have renewed the interest in the workings of the two main price setting markets: London's International Petroleum Exchange (IPE) and New York's Mercantile Exchange (NYMEX). The interaction of these two markets, when both of them are open (synchronous trading) and when only London is open (asynchronous trading), is important, in view of the fact that most participants take positions in both markets. This paper looks at how London is affected by New York by analysing the transaction duration of the IPE Brent futures contract, both when the NYMEX WTI futures contract is being traded and when NYMEX is closed. Using tick-by-tick data obtained from IPE, transaction durations are found to form two distinctive and inverted U-shaped patterns. Autoregressive conditional duration (ACD) model, first introduced by Engle and Russell, is applied to the data. Parameters of IPE morning and afternoon are significantly different from each other, underlining the dominant effects of NYMEX on IPE trading. The results from the current analysis reinforce previous results by the authors, which indicate that NYMEX is a leading price setter in crude oil futures prices and has a dominant effect on the IPE-traded contracts.

In the crude oil market, there are two ‘marker’ crudes that set the pace in prices: West Texas Intermediate and Brent Blend. The former is the base grade traded, as ‘light sweet crude’, on the New York Mercantile Exchange (NYMEX), while the latter is traded on London's International Petroleum Exchange (IPE) and is also one of the grades acceptable for delivery of the NYMEX contract. Participants in these markets move with relative ease from one market to the other and usually take positions in both of them. Arbitrage opportunities thus could be seized with and without the involvement of underlying physical crude. These arbitrage activities ensure deviations from the long-run path of two prices are adjusted and equilibrium prices are restored. Interesting questions thus arise: Which direction do arbitrages take place? Does one market move faster than the other? In other words, is there a market leader? To answer these questions, Engle and Granger (1987) cointegration analysis and vector error correction (ECM) model estimation are conducted.1 Both Brent and WTI daily futures settlement prices are integrated of order 1 and are co-integrated, which confirms the existence of a long-run equilibrium between the two markets. Further estimation with ECM reveals that short-term adjustments to the long-run cointegrating vector are insignificant in either market. Given that both markets stop trading at the same time, one possible explanation is that information released within the trading hours is fully incorporated in the settlement prices so that no lead-lag relationships can be detected using daily data. The search for the possible market leader leads us to experiment with higher frequency data: tick-by-tick duration data. This paper focuses on transaction durations of the IPE Brent futures contracts, both when IPE and NYMEX contracts are simultaneously traded and when NYMEX is closed, in order to uncover the effects of NYMEX trading on IPE. According to Easley and O’Hara (1992) timing between trades is related to new information. Analysing arrival times between events is a means of uncovering market information dissemination behaviour. Results of this paper uncover the high frequency facet of the markets. It is the first attempt, in our knowledge, to analyse transaction data in the area of energy futures markets. The rest of the paper is organized as follows. Section 1 introduces Engle and Russell's (1998) autoregressive conditional duration model (ACD). Section 2 discusses data construction and descriptive analysis. Section 3 presents empirical results and their implications and is followed by conclusions in Section 4.

This paper builds on previous work on the information transmission mechanism between IPE and NYMEX, with the crude oil futures contract providing the focal point. We concentrate on the intra-day behaviour of IPE prices and more specifically on trade durations. Empirical evidence suggests that the patterns of these durations are distinctively different between the morning and the afternoon IPE sections. This evidence suggests that NYMEX has a large impact on IPE, which was also the conclusion of our previous work. Whether this impact is the result of information disclosure, such as demand information, or simply due to the trade generated impact is an interesting topic for further study. As this paper goes to press, it is interesting to note that new developments on the physical side of Brent crude trading are afoot. The recent introduction of Brent–Flotta–Oseberg (BFO) composite crude illustrates the attempts to arrest the notable decrease in Brent crude reserves, and the possible repercussion such a decrease may have on the liquidity of the futures contract. Whether this move changes the behaviour of the IPE futures contract will be a matter for future research.