Denigrating democracy: How electoral competition fuels xenophobia in Lebanon.

When do electoral incentives encourage candidates to target immigrants with discriminatory rhetoric? I address this question through a novel dataset of refugee-related tweets by elites in Lebanon, the country hosting the most Syrian refugees per capita globally. I find that Lebanon's 2018 election, the first under a new set of electoral rules, precipitated an increase in anti-migrant tweets. However, the electoral campaign did not affect candidates' rhetoric uniformly, but rather fueled xenophobic discourses specifically among Christian candidates, whose voter base is particularly hostile towards refugees, and especially among Christians facing the fiercest electoral competition. This paper makes three contributions: theoretically, it elucidates the consequences of partisan competition for xenophobia; conceptually, it relaxes an assumption common in the ethnic institutions literature that ethnic composition is fixed over time; and empirically, it demonstrates how social media data can be harnessed for expanding a nascent literature on migration politics in the Global South. [ABSTRACT FROM AUTHOR]