The unexpected no to congestion charges

Abstract Public opposition is often regarded as the main political obstacle to implementing congestion pricing, yet it remains unclear why policies with concentrated private costs can face majority opposition. Using full-population registry data linked to referendum and general-election outcomes, we analyse the 2006 Stockholm and 2014 Gothenburg congestion-charge referendums. Car ownership is by far the strongest predictor of opposition, while political ideology explains additional variation beyond self-interest. Our main finding is that turnout changes the apparent importance of self-interest: districts with higher turnout show greater support, and controlling for turnout substantially strengthens the estimated effect of car ownership. The results suggest that referendum outcomes reflect not only who bears the costs of congestion pricing, but also who participates in the political process. JEL codes: R48; H23; D72; R41; D12