Debt and the Oil Industry - Analysis on the Firm and Production Level

Abstract

This paper analyzes the relationship between debt and the production decision of companies active in the exploration and production of oil and gas in the United States. Over the last couple of years, the development and application of innovative extraction methods, like hydraulic fracturing and horizontal drilling, led to a considerable increase in United States (US) oil production. In connection with these technological changes, another important economic development in the oil industry were largely debt-driven investments in the oil sector. The extensive use of debt was fostered by the macroeconomic environment of low interest rates and investors searching for yield in the aftermath of the financial crisis. Additionally, the rising prices in the commodities markets until mid 2014 led to higher asset valuation and thus to higher return expectations fueling a virtuous circle. This increased investment activity, especially in the US , raised the production capacity and as a consequence contributed to a higher production of oil and natural gas. This trend continued in spite of the oil price decline in 2014, whereas the oil price slump in 2008 led to a reduction in oil production, which seems to be the more plausible reaction. Using a novel data set the aim of this paper can be split into two research questions. The first research question is whether debt and leverage affects production decisions of companies active in the exploration and production of crude oil and natural gas. The second research question then is, if the technological changes in the industry and the increased indebtedness of US oil companies led to a markedly different reaction in their production decision following 2014 compared to the similar price decline in 2008. A potential reason for the absence or delay in cutting back production after the price drop in 2014 could be supposedly higher levels of debt prior to the price decline. These questions are addressed applying the relatively new panel vector autoregression approach to a novel dataset combining financial data on publicly listed firms and their production data on well level.