Abstract
Russian trolls generally supported the Trump campaign and were particularly active on Twitter 2015-2017. We find that trolling fell 35% on Russian holidays and to a lesser extent, when temperatures were cold in St. Petersburg. Exogenous variation in trolling by day allows us to considerindirectly-affected political behaviors in the US—outcomes that are less traceable via tweet sharing but potentially more important to policymakers than the direct dissemination previously studied. As a case in point, we describe reduced form evidence that Russian holidays affected daily trading prices in 2016 election betting markets. This response is consistent with successful Russian interference in support of Trump.

This publication has 21 references indexed in Scilit: