The influence of oil and natural gas employment on local retail spending: evidence from Oklahoma panel data

Recent volatility in oil and natural gas markets has led to questions about how local businesses and government revenues might be affected. This paper uses a panel dataset of Oklahoma counties during 2003–2012 to quantify the relationship between changes in oil and natural gas employment and retail spending. One particularly noteworthy contribution is the use of sub-categories of retail sales (to the two-digit Standard Industrial Classification code) to assess whether specific retail sectors are more responsive to changes in oil and gas activity. Our fixed effects panel regression model reveals that variations in mining employment impact aggregate retail expenditures, with an overall elasticity of 0.02. These results are driven by the grocery and furniture sectors, with elasticities of 0.05–0.06, and are even stronger when the analysis is limited to counties considered to be mining-dependent. We also show that spillover effects exist, with the grocery and furniture expenditures highly impacted by mining employment in neighboring counties as well as their own.