Aggregate Uncertainty in Runoff Elections

Coordination and Bandwagon Effects: How Past Rankings Shape the Behavior of Voters and Candidates

“…• Electoral commissions websites: We first checked whether governments make constituencylevel electoral results available through the websites of their electoral commission. This was the case for Bahrain 's 2002, 2006the Czech Republic's 1996, 1998, 1999, 2000, 2002, 2003, 2004, 2006, 2012, 2016, 2017Georgia's 2016 election;Lithuania's 2016 election;Mauritania's 2013 election;New Zealand's 1908 and1911 elections;Poland's 1989 election;and Switzerland's 1990and Switzerland's , 1991and Switzerland's , 1994and Switzerland's , 1995and Switzerland's , 1998and Switzerland's , 1999and Switzerland's , 2003and Switzerland's , 2007and Switzerland's , 2011and Switzerland's , and 2015 elections. • For countries in which some variables were missing, we supplemented our data with additional sources.…”

“…Most models of elections assume an exogenous pool of candidates. Models with endogenous candidate entry (Osborne and Slivinski, 1996;Besley and Coate, 1997;Solow, 2016;Dal Bo and Finan, 2018) and exit (Indridason, 2008) focus on individual candidates' choice of whether to run. In the real world, however, agreements between parties can also lead a candidate to drop out, thus restricting voters' options.…”

Majority runoff elections: Strategic voting and Duverger's hypothesis

“…This only ensures that a potential Condorcet loser never wins. 11 We argue that this result is closely related to the belief that the median candidate may be "squeezed" by a left-wing and a right-wing candidate (see, e.g., Solow 2013 andVan Der Straeten et al 2013) and, hence, excluded from the second round. We thus show that, perhaps surprisingly, such a squeezing can indeed happen when voters are strategic.…”

“…As mentioned in the Introduction, a common critique of the majority runoff system is that the median candidate may be "squeezed" by a left-wing and a right-wing candidate, and hence excluded from the second round (see, e.g., Van Der Straeten et al 2013 andSolow 2013). Arguably, such squeezing happened during the 2007 Presidential election in France (Spoon 2008).…”

“…8 By allowing for all possible preference orderings and intensities, we can consider any situation of interest as a particular case. For instance, combinations of policy platforms such that the median candidate is "squeezed" between two moderate candidates (see, e.g., Solow 2013, andVan Der Straeten et al 2013) or the majority is divided (see, e.g., Myerson and Weber 1993, Piketty 2000, and Myatt 2007 are easy to represent in our framework. Therefore, our results show how voters behave in these situations of particular interest in the literature, as well as any other that can come to mind (e.g., Osborne and Slivinski 1996, Callander 2005, and Bordignon et al 2013).…”