Terror and Conservative Shifts: Cross-fertilising approaches

Terrorist threats systematically shift political preferences toward right-wing positions, but existing evidence relies on either experiments with potential external validity concerns or quasi-experiments with limited mechanistic insights. We deploy complementary methodological approaches—a randomized video experiment in France and an event study linking German panel data to terrorist attacks—to establish both causal effects and underlying mechanisms. Despite fundamental differences in context, design, and treatment delivery, both methods estimate remarkably similar effects: approximately 5 percentage points increases in right-wing voting intentions, representing 9-14 percent rises relative to baseline support. Effects concentrate among older, less-educated, unemployed, and swing voters, and amplify substantially during electoral periods. The quasi-experimental design establishes external validity across two decades, multiple countries, and varying attack proximities, while demonstrating that even geographically distant incidents shift preferences. The experimental design illuminates mechanisms: voters respond specifically to security threats rather than equivalent-intensity environmental dangers, with anger and fear correlating strongly with political shifts. Immigration attitudes partially mediate these effects. Cross-fertilizing experimental and quasi-experimental evidence provides unusually credible causal inference while revealing that terrorist threats pose systematic vulnerabilities for democratic competition, particularly during campaigns when marginal voters finalize ballot decisions.