National Elections Across Democracy and Autocracy: Which Elections Can Be Lost?
with Susan Hyde from Yale University
Under review.
Abstract:
In some authoritarian regimes, election outcomes are determined with certainty. In others, established autocrats allow limited electoral competition, and a non-trivial number of incumbents have been surprised by election results, including the defeat of their own parties. Existing research makes clear that democracy does not necessarily precede contested elections, nor does democracy inevitably follow the introduction of electoral competition. The increase in “hybrid regimes” has produced a growing body of research on the determinants of electoral competition and its consequences in political, economic, and military affairs. Given that opposition parties can sometimes overcome election fraud, violence, and otherwise autocratic political institutions in order to win biased elections, how can researchers determine, ex ante, which elections can be lost? Which institutional characteristics divide non-competitive elections from those in which an opposition upset is possible? We argue that if opposition is allowed, multiple parties are permitted, and more than one candidate appears on the ballot, the minimal structural conditions for competition are present. We formalize this argument and introduce a new dataset that distinguishes elections that can be lost from those that cannot. We then outline the pitfalls of other methods used by scholars to define the potential for electoral competition, and show why such methods are likely to lead to biased findings regarding the causes and consequences of electoral competition.
Comments are closed.