Power Reliability and Regional Growth: A Double Machine-Learning Approach

Interruptions to electric power systems are to some degree inevitable. However, the frequency with which they occur, their duration, and the ability of providers to restore power quickly could have implications for regional growth. Using information on distribution utilities and power marketers of electricity across the U.S., we examine the relationship between grid reliability and county-level growth. Specifically, we utilize a double machine-learning technique to assess how various measures of power reliability are associated with the percent change in employment and population. Overall, the results suggest that counties benefit from improvements in power reliability. We also find linkages between shorter minor interruptions and regional growth, especially in rural areas. JEL Classification : R10, Q40, C10