We provide evidence that violence reduces the adoption and use of mobile money in three separate empirical settings in Afghanistan. First, analyzing nationwide mobile money transaction logs, we find that users exposed to violence reduce use of mobile money.