Why do states join wartime coalitions despite the absence of a salient threat or strong ties to the coalition leader? We argue states make unexpectedly high contributions to coalition warfare as a costly signal of their desire for a stronger relationship with the coalition leader. Conventional theories insufficiently explain why states without immediate security interests or strong ties to the lead state over-contribute relative to their capacity. Using newly compiled data on troop contributions to the war in Afghanistan (2001–2014), we find states are most likely to contribute a higher share of their armed forces when their relationship with the US has unrealized alliance potential. States with under-performing alignments leave substantial room for subsequent gains to be had from signaling their commitment to the leading coalition actor. Our finding helps explain why states risk the costs of war—casualties and domestic accountability—by participating in coalition warfare.

An American’s yearly chance of being killed by a terrorist attack sits at roughly 1 in 3.5 million. Yet, over 40 percent of Americans consistently believe that they or their family members are likely to be a terror victim. Can these inflated estimates of the risks of terrorism be brought closer to reality? With trillions of dollars spent on the “War on Terror,” this question is not just theoretically but practically important. In order to investigate, we use an experimental approach assessing whether people update their beliefs about terrorism when given factual information about the relative risks it presents. We find that public fear of terrorism and demand for countering it can be sharply reduced with better information, dropping essentially to pre-9/11 levels after the treatment and staying that way two weeks later. These results suggest that countering the indirect costs of terrorism may largely require providing more context and perspective.