The events of the past few weeks in Belarus have been deeply inspiring, as tens of thousands have turned to the streets to demand clean elections, despite brutal crackdowns by the security forces. Despite mass arrests and evidence of beatings and torture in the detention centers, the size and scope of the protests have only grown.
As someone who is committed to liberal democracy, I can only be inspired by the courage of the citizens, and the inventiveness of the opposition. Before the election, for example, the opposition crashed an authoritized pro-government rally to get around a ban on their own gatherings.
As a social scientist, though, the protests highlight one of the most important puzzles in the study of manipulation elections, and–in my view–of the democracy and authoritarianism broadly:
Why does post-election protest happen?
Mass protests in support of clean election make headlines when they happen, in part because they are actually quite rare. From 1991 to 2009, Kalandadze and Orenstein count just 17 cases of mass electoral protest. This is despite the fact that low-integrity elections take place regularly across much of the globe.
This represents something of a puzzle, since the prevailing models of election manipulation assume that mass protest is the most important deterrent to electoral malfeasance by governments. These are generally rational choice models of manipulation, in which the opposition player chooses whether or not to protest based on its strategic calculation (See for example Magaloni 2010 and Little 2012). In some of these models, the larger the amount of fraud, the more likely protest is to occur (e.g. Fearon 2011). But this expectation clashes with the results in Alberto Simpser’s pathbreaking book, which argues that elections characterized by blatant and excessive manipulation are actually less likely to spark protest than those where manipulation is marginal. So where does that leave us?
Events in Belarus and a research agenda for understanding electoral protest
How does our understanding of electoral protest help us understand the events in Belarus, and what can they tell us about how to move forward? Perhaps the most significant feature of the situation in Belarus that conforms to the social-science literature is the decision by the major potential opposition presidential candidates to coalesce around Svetlana Tikhanovsksaya prior to the election. This was in part perhaps an unforced error by the regime, which chose to exclude other major contenders from the ballot (including Tikhanovskaya’s husband, anti-regime blogger Sergei Tikhanovsky). By excluding all credible candidates but one, the regime created an opportunity for the kind of opposition coalition that has been associated with liberalizing outcomes (Howard and Roessler 2006), or at least an opposition victory (Wahman 2013). Tikhanovskaya, in exile in Lithuania after being forced to read a conciliatory statement from within the Central Election Commission, has claimed about 70% of the true vote; the CEC gives her about 10%. The true figure may be closer to a tie in the mid-40s (Rus) between her and incumbent president Lukashenko.
Another important element to consider is the way in which the election was manipulated. Anecdotal observer reports and more systematic citizen observations suggest that widespread falsification–rather than vote-buying or voter pressure–was the order of the day. In a recent paper with my coauthor Paula Mukherjee using cross-national data from 1980 to 2004, we find that elections characterized by falsification are more likely to result in post-election protest than those where manipulation is carried out in more costly ways.
As the above figure shows, elections with widespread fraud are more likely to generate protest than those which include both fraud and costly efforts at mobilization. This seems to have been the case in Belarus’s 2020 election.
Finally, events like the protests in Belarus highlight our need to better understand how ordinary people think about election integrity, and how movement entrepreneurs try to shape that thinking. Why did Lukashenko’s ostensible 80% victory spark a nationwide convulsion in 2020, but his 84% victory in 2015 did not? To answer questions like this, I think we need to turn to social psychology; to questions about how election manipulation drives feelings of anger, moral indignation, or group identity. In my view, this is an open frontier for research on election manipulation, with the potential to improve our understanding how ‘self-enforcing’ democracy really is. To that end, I will be conducting survey experiments around the upcoming US election, with the aim of illuminating how changes in the nature of election manipulation affect these protest triggers. My hope is that research in this area will help us explain why post-election protest occurs only rarely, and how pro-democracy advocates can best appeal to the people to defend fair elections.