Developing a hierarchical system for energy corporate risk factors based on textual risk disclosures

Abstract

Selecting risk factors is essential for measuring energy corporate risk. However, the comprehensive identification of energy corporate risk factors is still a difficult issue. This paper innovatively uses the text mining approach to comprehensively identify energy corporate risk factors from textual risk disclosures reported in financial statements. Based on 131,755 risk factor headings from 3707 Form 10-K filings from 840 U.S. energy corporations over the period 2010–2016, 66 types of risk factors that affect energy corporate risks are identified. Furthermore, we develop a hierarchical system for 66 energy corporate risk factors by dividing energy corporations into nine subsectors. Thus, the hierarchical energy corporate risk factor system provides fundamental support for further energy corporate risk measurement. Researchers can comprehensively and effectively select risk factors in measuring risks of the entire energy industry or each of nine energy subsectors.