"This paper explores the problem of spillover bias in cross-border studies of the minimum wage using a commuter gravity model on county-level data from 2009 to 2013. Commuter flows conform to the expectations of the gravity equation, but flows across county borders are sensitive to changes in the minimum wage rate, which implies that minimum wage employment effect estimates using contiguous counties are likely to suffer from spillover bias. One way to address this bias is to include a control ring between treatment and comparison counties, although this solution may introduce biases of its own. A gravity model that includes a control ring affirms that this alternative comparison group can address this spillover bias problem."

Support: Digital