Working papers

Inglorious wars and political radicalisation

Abstract. This paper investigates how forced conscription in war periods affects post-war political behaviour. During WWII men from Eastern France were forcibly conscripted to the German army. A discontinuity in the draft rule within affected regions allows establishing a plausibly causal link from this conscription to post-war voting behaviour. Municipality-level evidence indicates that past incorporation has a positive and significant effect on abstention and support for anti-establishment candidates in post-War elections. To test whether this behaviour is the result of low political trust I exploit content analysis and survey data. Wehrmacht conscription increases electoral support for candidates that are more critical towards the polity. Moreover, individuals living in places more exposed are more critical of political classes and less trustful of institutions.

Keywords. War, political trust, political radicalisation, persistence

JEL classification. D72, N44, F51, Z13

  • gallery-image
  • gallery-image
  • gallery-image
[unex_ce_button id="content_94t1d5tbs" button_text_color="#ffffff" button_font="semibold" button_font_size="15px" button_width="auto" button_alignment="center" button_text_spacing="2px" button_bg_color="#000000" button_padding="10px 15px 10px 15px" button_border_width="2px" button_border_color="#000000" button_border_radius="0px" button_text_hover_color="#ffffff" button_text_spacing_hover="2px" button_bg_hover_color="#000000" button_border_hover_color="#000000" button_link="" button_link_type="url" button_link_target="_blank" has_container="1" in_column=""]Download paper[/ce_button]

THE LOGIC OF FEAR - Populism & Media Coverage of Immigrants Crimes -

with M. Couttenier, S. Hatte, and M. Thoenig

Abstract. We study how news coverage of immigrants’ criminality impacted municipality-level votes in the November 2009 referendum on "Minaret Ban" in Switzerland. The campaign, successfully led by the populist party SVP, played aggressively on the fears of Muslim immigration and linked Islam with terrorism and violence. We make use of an exhaustive dataset of violent crime detection that we combine with detailed information on crime coverage from 12 newspapers. First, we quantify the extent of the pre-vote media bias in covering migrant’s criminality. Second, we estimate a theory-based voting equation in the cross-section of municipalities. Exploiting random variations in crime occurrences, we find a positive and large effect of news coverage on the votes in favor of the Minaret ban. Our counterfactual quantifications show that, in presence of a law forbidding the report of crime perpetrator nationality, the outcome of the referendum would have been nearly reverted at the national level (from 58% to 51%).

Keywords. Media, violent crime, immigration, vote, populism

JEL classification. D72, L82, Z12, K42

  • gallery-image
  • gallery-image
  • gallery-image

Work in progress

Technological innovations in electoral campaigns: Direct canvassing and partisan mobilization

with J. Buggle

Abstract. Despite the opportunities that technological innovations offer to reach an ever-growing audience, traditional means to access the electorate, such as direct canvassing, are widely used. In this paper we argue that the main reason direct canvassing persists as a campaign strategy is because it is a very efficient mean to mobilize voters that are already supportive of a candidate's program. To test our hypothesis, we exploit a unique historical context: in 1896, due to limited funding, the Democratic Presidential candidate adopted an unprecedented campaign strategy by using the railroad to go on a national speaking tour. We make use of the railroad network to draw causal inference of the impact of direct canvassing on the electoral outcome. Results indicate that the Democratic candidate obtained significantly more votes in places where he gave a speech. We then exploit political competition in the House of Representatives election to disentangle between votes gained from persuasion and mobilization. The electoral gain of the Democratic candidate can be decomposed into 65% from increased mobilization of his electorate, and 35% from persuading Republicans.

Keywords. Electoral campaigns, mobilization, persuasion, political competition

JEL classification. D72, N41, N71, P48

The long-run consequences of extractive institutions: Economic Aryanization

with J. Senn

Abstract. This paper investigates the channels through which extractive institutions affect development. We test the hypothesis that extractive institutions reduce competition, which in turn reduces the incentives of firms to innovate, and eventually slows down growth (Acemoglu and Robinson, 2012). During the German occupation of France in WWII, the policy of «Economic Aryanazation» was introduced. The purpose of this policy was to exclude the Jewish population from any economic activity. Variation in this set-up comes from the «Demarcation line», the line separating France into the Occupied (North-West), and «Free» (SE) zones for 2.5 years. The initial drawing of this line was so imprecise, that the authorities of the «Free» zone had to wait several months before learning where the border was fixed (Alary, 2003).