“…Additionally, we have to exclude potential spillovers from treated regions to nontreated ones as these would invalidate the identifying assumptions of SCM. We do this, as is standard in the regional policy evaluation literature (Alder, Shao, & Zilibotti, ; Aragon & Rud, ; S. O. Becker, Egger, & von Ehrlich, ; Dettmann, Brachert, & Titze, ), by considering spillovers to be a function of proximity to the treated locations and, hence, exclude regions neighboring World Cup venues (and Lesotho) from the donor pool. Under the hypothesis of no indirect effects on nontreated regions, using a donor pool stemming from the same country as the treated unit should be a first best solution, since such a donor pool offers better comparability than a donor pool from any other country .…”