Bootleggers Behind the Baptists

Wine-Industry Influence in Switzerland’s 1908 Absinthe Referendum

Did wine producers — not just public-health reformers — drive Switzerland’s 1908 vote to ban absinthe? A ‘Bootleggers and Baptists’ test using a newly assembled canton-level dataset built around a previously unanalyzed 1907 enumeration of the absinthe industry.

Working paper (preliminary) · with Nicholas A. Jensen (Wabash College)

Overview

On 5 July 1908, Switzerland became the first country to ban absinthe by direct popular vote — passing the constitutional ban with 63.5% support, over the Federal Council’s recommendation to reject it. The campaign was led by public-health and temperance reformers, the “Baptists.” But absinthe had grown popular during the phylloxera years precisely as a cheap substitute for wine, putting it in direct competition with Switzerland’s wine producers. This paper asks whether those producers — the “Bootleggers” — quietly moved the cross-canton vote, using a 25-canton dataset built around a previously unanalyzed firm-level enumeration of the Swiss absinthe industry compiled in 1907 for the Federal Council. It is part of my broader interest in how public-health rhetoric can mask economic self-interest, especially when protectionism is enacted directly by voters.

Abstract

“On 5 July 1908 Swiss voters approved a constitutional ban on absinthe by 63.5 percent — the second European country to prohibit it and the first by direct popular ballot. We test whether wine producers — the upstream category whose displaced demand absinthe had absorbed during the phylloxera years — moved the cross-canton vote, using a 25-canton dataset built around a previously un-analyzed firm-level enumeration of the Swiss absinthe industry compiled in 1907 for the Federal Council. The raw wine–Yes correlation is negative; conditioning on French-speaking share flips its sign, the textbook Simpson’s-paradox reversal. Fractional-logit average marginal effects then show that a one-percentage-point increase in a canton’s national wine-revenue share raises its yes-share by 0.43 percentage points; for the dominant wine canton, Vaud, this implies roughly 15 percentage points of its predicted yes-share. The wine signal is absent from the 1906–07 petition stage and from a same-day commerce-vote placebo: a Bootlegger coalition that consolidated at the vote rather than at the petition.”

Note

This is a preliminary working draft — please do not cite or circulate without permission. A current version is available on request.