Association for Public Policy Analysis & Management

The fracking boom over the past decade has transformed many regions of the country as oil and gas valued at billions of dollars has been extracted from shale plays across the United States. Many of these areas have seen an increase in overall economic activities and new wealth from oil royalties. However, in some states and local communities there has been pushback to allowing future hydraulic fracturing activities. This is in part due to the drastic increase in crime during this period in fracking areas. However, the causal drivers of this change are unknown, especially since local aggregate crime patterns are not able to account for migratory workers.

This paper addresses two questions concerning criminal behavior in fracking communities using all criminal cases filed in the state of North Dakota, the second largest oil producing state, from 2000-2017. First, using a census of all rural residents prior to the fracking boom, I am able to separate crimes committed by residents and those by transitory workers. To identify the overall effects of fracking on residents, I compare residents living in the Bakken shale play to those living off of the shale play, before and after fracturing activities began. Second, I examine the wealth effect of fracking on criminal behavior, as income is a known correlate of crime. As not all landowners retain their mineral rights or have oil beneath their property, some households experience large income shocks via royalty payments while others do not. To isolate the effects of increased financial resources within a household, I exploit within- and across-county variation in mineral rights ownership of landowners along with the timing of royalty payments.

Using data collected on all residents, leaseholders and criminal cases filed in North Dakota along with the above empirical strategies, these findings will speak to the causal effects of fracking in local communities and the mechanisms behind them. Moreover, as not all crimes are alike, I am able to detect heterogeneous effects by crime category. In doing so, these findings contribute to a broader literature on how changes in income affect crime.