Joshua Tucker: The following guest post by political scientists Emilie M. Hafner-Burton (University of California at San Diego) and Susan D. Hyde (Yale University) is part of our ongoing collaboration with political science journals to bring you summaries of recently published work and to make that work available for free download for a period of time. This post is related to an article recently published by Hafner-Burton, Hyde, and Ryan Jablonski in the British Journal of Political Science entitled “When Do Governments Resort to Election Violence?” and which can be freely downloaded here.
Democracy and human rights are supposed to go hand in hand. International law establishes that the authority to govern should be based on the will of people to express themselves through genuine elections. In fact, all but a few governments in the world now organize regular elections for national office. Yet the reality is that elections are sometimes violent and dangerous events, and the government is often to blame for instigating a cycle of election violence. In order to stay in power while facing electoral challenges, some governments resort to physical violence and intimidation against citizens, opposition candidates and political parties. Some governments use violence before the election to win; others use violence after the election to put down peaceful post-election protests.
There are lots of examples. Election violence has worked well for the ruling New Azerbaijan Party, which routinely arrests journalists and activists, tries to prevent the opposition from campaigning and sends security forces to beat people who dare to protest election results. It worked in Iran, when — a few years ago — the ruling government violently put down mass protests sparked by the presidential election; many people were killed. And violence is commonplace in Bangladesh, where general elections are routinely marred by extreme violence on the part of the government and opposition parties. The conventional wisdom is that leaders resort to violence because it helps them win them elections and — afterwards — to stay in office if the election results are challenged by popular protest. Yet not all leaders who have the means and the motivation to use election violence actually do so. History shows a tremendous amount of variation in when and where leaders use such tactics. Why do leaders sometimes resort to election violence and sometimes exercise restraint?
Our research, co-authored with Ryan Jablonski (and temporarily ungated in the British Journal of Political Science), suggests a relatively simple answer. Leaders are more likely to crack down when they think an election might unseat them (or their party) from power and they face few constraints on their authority and so have reason to believe they can get away with violence. The most repressive and least democratic regimes in the world have no need for election violence. In North Korea, real electoral competition has historically been impossible, and election-specific violence is unnecessary to ensure a favorable election outcome. No opposition candidates ever appear on the ballot, so the risk of losing an election is non-existent. In other places, the government may face an electoral threat but does not respond with violence because leaders are more likely to be held accountable. When judiciaries become more independent of the executive office, or when other checks on government power develop, election violence becomes less likely even when a leader or party’s position in power is seriously threatened.
Although it may seem obvious to say that leaders use election violence as a way to fend off threats to their power, it is not at all obvious how to identify these threats before the violence breaks out. When do leaders feel threatened by elections? And when do they feel immune to punishment? There is no direct way to measure what’s going on inside the head of leaders like President Mugabe of Zimbabwe when he faces elections or their aftermath—for instance, whether he worries about losing or is confident of a genuine victory. But there are clues. One clue comes from the existence and content of public opinion polling about the election. In the period leading up to the election, some leaders can get information about their popularity from credible public opinion polls. A surprising 62 percent of elections in our study (from 1981-2004) had them. Such polls are one way to gauge whether or not a leader faces an electoral threat. Another clue can be found in candidates’ pre-election statements about whether they expect to win. Looking to history, these clues seem to have predictive power. Among leaders that had few constraints on their authority, those that were uncertain about their own victory were significantly more likely to use violence leading up to an election than leaders who appeared confident they would win.
These findings have some important implications. Countries that are most likely to experience election violence are precisely those places in which the government feels threatened by its opposition. Although repression is clearly detrimental to democracy in the short term, competition is important for democracy. If our analysis is correct, then election violence may be a symptom of a threatened and potentially weakening incumbent government rather than a sign that democratization – and future protection for human rights – is doomed.