Casinos as an Economic Development Strategy in the Prairie Provinces of Canada

Using data from several census periods, we examine whether the establishment of casinos affects employment in the Prairie Provinces of Canada. Using a standard two-way fixed effects (TWFE) estimator, we find some limited evidence that census subdivisions (CSDs) that opened casinos experienced increases in employment. However, given recent research questioning the robustness of TWFE results in circumstances similar to ours, we use a multi-period Callaway and Sant'Anna difference-in-differences (CSDID) module and find that the estimated employment effects are much reduced. Furthermore, a key hypothesis that employment in CSDs with and without casinos evolved similarly before the establishment of casinos is rejected. This parallel-trends assumption is required to infer a causal relationship regarding the establishment of casinos and changes in employment. A central issue that arises is whether the five-year census period is sufficient to determine whether CSDs with and without casinos have the same event study dynamics. We also find that identifying the benefits of First Nations casinos for First Nations people is challenging because urban reserves do not constitute a separate CSD.