Tracking Disruptions

Bauer Researcher Uses Big Data to Track Oil Refinery Disruptions

Published on March 14, 2022

Photo: Bin Li

New research from the C. T. Bauer College of Business uses big data to track operational disruptions at U.S. oil refineries that cause significant reductions in production capacity. The research offers important evidence that is likely to be of interest to regulatory agencies, investors, and consumers, because the disruptions impact petroleum product prices and stock prices but largely go unreported.

“If all the refineries near a particular city experience an outage, local gas prices will increase by 8.7 cents per gallon the day after the production outage,” the researchers write. Overall, this price effect can be translated to a $26 increase in the quarterly cost of gas per household. Given the significant price impact of refinery outages, this evidence should be of interest to U.S. regulatory agencies, which have discussed mandating firm disclosures of refinery outages for more than a decade.

Assistant Professor Bin Li of the Department of Accountancy & Taxation co-authored the study, which is forthcoming at Journal of Accounting Research, a top Accounting journal.

The research, “Leveraging Big Data to Study Information Dissemination of Material Firm Events,” expands what is known about the frequency and scale of refinery outages by utilizing the power of high-frequency cell phone “pings” (i.e., geolocation signals from mobile devices) to track unusually high foot traffic that occurs as plants bring in support crews to get refineries back in operation. The study takes into account cell phone signals that appeared at almost all (more than 85 percent) major U.S. oil refineries.

“Our analysis shows that refining firms do not voluntarily disclose outages, and that traditional media cover only about one-third of outages,” Li said. According to Li’s research, traditional media report just 27% of the monthly offline events at refineries, on average. Furthermore, firms do not report such outages in Form 8-K SEC filing either. “The lack of firm disclosures about outages should be helpful to the Securities Exchange Commission in evaluating firms’ compliance with 8-K filing requirements,” the researchers write. How then will investors get this information? “Investors appear to learn the financial impact of such outages through subsequent earnings announcements,” the paper states.

According to the findings, traditional media appear to capture more news about planned than unplanned outages, report relatively more refinery outages in big cities, cover refinery outages of publicly traded companies, and report outages in refineries with higher levels of production capacity.

Stock returns do not reflect the implications of refinery outages until about three to four months after the outage. So, there is room for investors to profit from this delayed stock market response by using big data.

About Professor Li: Li’s research concerns whether and why accounting disclosures, both mandatory and voluntary, affect investor decisions and firm value. He is also interested in research topics related to corporate finance, regulatory enforcement and standard setting. His research has been published in top accounting and finance journals and has received multiple awards, such as the Notable Contributions to Accounting Literature Award and the FARS Best Paper Award. Prior to joining the Bauer College, Li was an Assistant Professor of Accounting at University of Texas at Dallas and University of Oklahoma. He received his Ph.D. in Accounting from Duke University in 2012.