Effects of the shale boom on ethylene and propylene prices

ABSTRACT

This study analyzed the effect of the shale boom on the olefin industries. We investigated the paths by which US natural gas prices are transferred to ethylene and propylene prices and identified the price determinants based on the supply and demand conditions of each commodity’s market. By comparing with the Japanese markets, the understanding of the supply structure of the ethylene and propylene industries is broadened. A path analysis framework was constructed to reflect the supply chains and applied to 2009–2018 data. Results showed that the downward trend in ethane prices provoked by the US shale boom reduced ethylene prices and raised propylene prices. Propane still serves as a determinant of the latter, while naphtha has lost predictive power for both. The comparative data confirmed that the US produces most of its petrochemicals from natural gas and gas-based feedstocks, while Japan relies on crude oil and refined naphtha.

KEYWORDS:

• Petrochemicals
• ethylene
• propylene
• ethane
• shale gas
• supply chain
• path analysis
• structural equation model

Disclosure Statement

All other authors report no conflicts of interest relevant to this article.

Notes

¹ EIA analyst Warren Wilczewski stated in a presentation that “the only data that will no longer appear in EIA tables is ethylene, with the exception of the refinery production of ethylene.”

² Since our primary goal is to investigate the effect of crude oil and natural gas prices on ethylene and propylene prices, the paths from stocks to prices of crude oil and natural gas are excluded to start our analysis from the prices.

³ In the analysis, the quantity demanded is assumed not to have changed relative to the quantity supplied, since domestic demand and exports of ethylene (both kinds of data are not obtainable) are not considered to have increased dramatically. The main source of domestic demand for ethylene in the US is in polyethylene production facilities, but the growth rates of polyethylene capacity and production are known to be slower than those for ethylene (Waldheim 2019). While ethylene export facilities and exports are forecasted to increase, the facilities were under construction during the analysis period (TCG 2017). In addition, as ethylene exports in 2017 were 98.9 million dollars (1.3% of the global exports) and 97.9M million dollars in 2010 (OEC 2018), it is difficult to say they increased significantly over the period from 2010 to 2018. Along with these factors, as market participants have focused more on ethylene supply as a major price determinant than on export and consumption, we assume that the ethylene supply alone can represent market tightness in terms of quantity, on the presumption of quantity demanded (consumption and exports) being unchanged.

⁴ The single covariance over the entire period between natural gas and crude oil prices may be a strict assumption, given some evidence that their relationship has shifted. For example, Brigida (2014) found that the relationship regime-switches over time, and Hartley and Medlock III (2014) showed that the technology change in electricity and exchange rates are attributed to the shift in the oil and gas relationship. In addition, Roberts (2019) provided statistical evidence that the cointegration of oil and natural gas prices disappeared after the previous shale boom. These time-varying relationships can be considered using the dynamic structural equation model of Asparouhov, Hamaker, and Muthén (2018), but this study excluded these time-variances from consideration to simplify complex path analysis and to secure degrees of freedom from our limited observations.

⁵ Standard regression models analyze unit-level relationships, and thus ignore the effect from explanatory variables being transferred to explanatory variables through mediating variables.