Drilling Down the Bakken Learning Curve

Improvements in horizontal drilling have helped unlock U.S. tight oil plays and reverse the decades-long decline in domestic onshore oil production. Whether low oil prices will turn the shale boom into a bust depends in part on how companies have reduced the cost of drilling wells. This paper investigates the role of learning-by-doing in drilling horizontal wells in the Bakken Shale Play. I use a large set of data on oil wells drilled in North Dakota between 2005 to 2014 to measure how firms reduce drilling times as they acquire experience. The results show that as firms gain experience in the Bakken, they drill wells faster. A doubling of a drilling rig's experience leads to a 5% reduction in the time to drill a well, which translates into a cost savings of about $31,000 per well. Given that on average a rig drills eight wells per year in the Bakken, a rig is expected to reduce the time it takes to drill a well by 11% over its first year of drilling. These findings have implications for how the current low oil price environment will affect U.S. drilling activity. Furthermore, there is evidence of organizational forgetting by rigs resulting from breaks in between drilling wells, which suggests productivity will be negatively impacted if and when drilling activity rebounds. Lastly, I find no evidence of learning spillovers across firms. This result implies that firms internalize the knowledge gained that is relevant for reducing drilling times in subsequent wells, and it has implications for social welfare and the use of fossil fuel subsidies.