July 19, 2010, 1:03 p.m.
Elections in emerging democracies are increasingly important components of efforts to manage conflict as well as facilitate political reform and economic growth. As a result, the international community rightly continues to support electoral processes, both diplomatically and by lending significant technical resources.
However, electoral fraud continues to bedevil the management of these elections, producing large-scale protest and violence of the kind recently seen in Iran, Nigeria, Ethiopia, Kenya, and Zimbabwe. Fraud also threatens citizen perceptions of government legitimacy, making it less likely that people think it is worth their time to turn out and vote in the future.
While the international community continues to bolster electoral processes through democracy and governance programs, we lack a firm understanding of the causes of fraud and its effects. For example, although elections in Sierra Leone and Liberia have helped to legitimize democratic governments and further reconciliation among ex-combatants, recent nationwide elections in Iran, Nigeria, Kenya, and Zimbabwe suffered malfeasance and vitiated democratic consolidation.
Why do elections sometimes facilitate democratization while at other times subvert it? Many academics, journalists, and policymakers have argued that citizens in these countries may lack democratic values or commitment to fair elections. Moreover, ethnic diversity is thought to exacerbate group conflict where politicians treat elections as winner-take-all contests and therefore have large incentives to rig. If either is the case, there is probably little that donors can do to intervene and support legitimate elections.
However, democratic culture and ethnic diversity do poor jobs of explaining variation among electoral outcomes. Take the cases of recent elections in Ghana and Kenya. Surveys in both countries show similar levels of support for democracy, and each country contains a number of ethnic groups within its borders with a history of inter-group conflict. Although the outcomes of both elections were razor thin, a clean election helped produce a peaceful incumbent party turnover in Ghana whereas rigging resulted in mass violence in Kenya. I suggest why this was the case below.
Our lack of clarity on the causal links between elections and violence means that policymakers may not have the information they need to ensure that elections help solve rather than instigate conflict. This information is vital because post hoc solutions to electoral violence (power-sharing, for example) have not proven fruitful and, in many cases, have been destabilizing.
In Kenya and Zimbabwe, coalition governments have not brought either desired reforms or reconciliation between parties that the international community had advocated in supporting these processes. Power-sharing solutions were suggested as possible avenues to peace after disputed results in Afghanistan and Iran, but the expectation of coalition government between parties that have fought hard campaigns and hold divergent policy preferences may be unrealistic and un-workable.
Moreover, similar to rigging, power-sharing abrogates the democratic will expressed by citizens who turned out to vote. No one would have considered executive power-sharing between George W. Bush and Al Gore after the Florida re-count as a "democratic" solution to contested results, so we should not hold different expectations of citizens in new democracies.
Given these problems, the policy puzzle is how to ensure the proper conduct of elections in the first instance.
Towards that end, the international community should re-focus on the important institutional components that are required for free and fair electoral processes. Not surprisingly, this includes the creation and maintenance of independent electoral commissions. Given that all parties to an election may have incentives to cheat - especially when polling reveals that races are close - meaningfully independent and robust commissions serve as important third party guarantors that alleviate credibility problems between parties otherwise willing to resolve their disputes through extralegal means. Without this check, the uncertainty of outcomes in the shadow of elections makes all sides more likely to rig the process and spark violence.
Specifically, the international community can support fair elections in a number of ways. First, research by Susan Hyde, a professor at Yale University, shows that international observers can offset incentives to cheat by observing the voting process at polling centers. This is an important and established activity, but given the systematic rigging that can occur within an election commission, observers must also be present for the lifetime of the tallying process — even if the count takes weeks or months, as it did for Afghanistan and Iraq’s last elections.
Second, monitors should consistently and systematically employ independent vote count verifications to assist commissions and lend support to the final tally of votes, including parallel vote tabulations (PVTs) and exit polls. PVTs are scientifically sampled and independently verified counts of actual ballots and can be easily reported and relayed by observers with cell phone photo capture and text messaging. Because PVTs cannot detect ballot manipulation, exit polls are an important check on results that gauge voter intentions independently of possible attempts at burning or stuffing or other problems with voting technology (whether intentional or otherwise). As an added benefit, exit polls also help researchers probe the determinants of voting behavior, an under-studied aspect of elections in new democracies.
In clean elections, projections of winners produced by PVTs and exit polls should closely approximate the announced results within small margins of error. When they do not, fraud has most likely occurred. Neither PVTs nor exit polls are new, and they have been employed individually in many cases. They also have some individual drawbacks, but when used in conjunction, they are important complements to each other and should become required components of international monitoring missions, even in countries that have run clean elections in the past. Their results should be released in a timely manner to lend credence (or question) to tallies produced by commissions. PVTs and exit polls are also relatively cheap and more effective than other forms of democracy assistance.
Third, the international community should identify and buttress domestic observers and civil society actors who are capable of defending commissions’ independence and mediating disputes between political elites should they arise over voting and the count. Recent research that I have done with Barak Hoffman and Katey Metzroth of Georgetown University demonstrates the importance of having independent and robust civil society monitors in helping to support the work of Ghana’s electoral commission and reduce fraud. Working with government and party elites, civil society leaders in that case were able to dampen political tensions and ensure a peaceful party turnover in a hotly contested and narrowly won race.
Given the multitude of problems associated with direct foreign involvement in negotiations after disputed outcomes, these local civil society actors may be more successful at ensuring peace and also bring important local ownership to electoral management. However, the international community will need to help organize and coordinate leaders and groups, as well as ensure that they are credible. Problems of coordination and independence bedeviled domestic monitoring efforts in Kenya. Coordinating civil society may not always prove easy or straight-forward, but it should be a vital aim of long-term observation missions.
Fourth, an important corollary of supporting civil society should be the deployment of a robust network of nation-wide domestic observers in the months before an election who report on human rights violations and violence before, during, and after electoral processes. Such data can be gathered systematically and mapped through Ushahidi’s crisis crowd-sourcing technology. This will provide civil society groups and monitors real-time data on hot spots to allow for quicker interventions to prevent further violence.
Given their commitment to governance reform and the promotion of democracy world-wide, Western donors can lend assistance to employ these simple techniques consistently across cases to increase the probability of legitimate elections and therefore reduce the likelihood of fraud and post-election violence. This no doubt serves the policy interests of the international community, but also the desire of citizens in emerging democracies to participate in free and fair elections.
James D. Long is a PhD candidate in Political Science at the University of California, San Diego and a 2009-10 Jennings Randolph Peace Scholar at the U.S. Institute of Peace. He writes on public opinion, voting behavior, electoral fraud, and violence in the developing world and has observed elections in Kenya, Ghana, and Afghanistan.