Markets and Democracy Briefs are published by CFR’s Civil Society, Markets, and Democracy initiative. They are designed to offer readers a concise snapshot of current thinking on critical issues surrounding democracy and economic development in the world today.
From Egypt to Russia, international election observers are nearly everywhere these days. Even Myanmar has, for the first time ever, invited international observers to the upcoming elections and, with just a week to go, the U.S. State Department has accepted the invitation. But is all this monitoring a good idea? When election monitoring first took hold in the late 1980s, opinions ranged from enthusiasm to skepticism. In those days, there was little evidence as to whether election observation really worked or how best to conduct it. Over the years, however, this has changed: election observation has become a dominant tool in the promotion of democracy. But the acceptance of monitors’ pronouncements has become too routine: we need to understand the nature of the various organizations that participate in monitoring, as well as its limits and even its risks. And we need to continue to seek improvements in its practice.
The Power of Election Monitoring
Election monitoring can improve the quality of elections and make incumbent turnover more likely. Research on election monitoring worldwide from 1980 to 2004 shows that in multiparty states that were not yet fully established democracies, monitored elections were 10 to 20 percent more likely to be seen as representative. Furthermore, incumbents lost power more often than in non-monitored elections.1 This does not prove that monitors caused these improvements, but the analysis did account for various factors to make sure the findings were not simply due to a tendency to observe elections that were in any case likely to be better. This makes it more likely that monitors had something to do with improved outcomes.
Indeed, countries respond to monitors’ advice if they want to improve their reputation and gain the favor of the international community, or if they are reform-minded but lack experience with democratic elections. For example, in response to monitors’ recommendations, Albania reformed its electoral laws, El Salvador improved voter lists and ID cards, Indonesia changed how its ballots are marked, Mexico rewrote campaign finance laws, and so on. Some of these improvements also had domestic advocates, but monitors brought valuable attention to the issues. Over the years, in countries like Mexico and Egypt, international monitors have also trained domestic monitors. Famously, in Panama, monitors worked with the Catholic Church to organize a parallel vote count, which helped discredit the fraudulent 1989 election.
International monitors can also provide reassurance after hotly contested elections. In South Korea in 1987, Bulgaria in 1990, and Mozambique and Mexico in 1994, the environment was so polarized that without international monitors the victor might not have been able to establish a governing mandate. As exemplified by former president Jimmy Carter and his Carter Center, monitors can also play crucial postelection mediation roles.
This is all good, and it points to why election observation is here to stay.
The Limits of Election Monitoring
However, as the practice of international election observation is becoming ubiquitous, it has become evident that it doesn’t work everywhere: research has found that even when monitors are present, politicians cheat in obvious ways nearly 17 percent of the time. Another 24 percent of elections fall in a gray zone, meaning that serious problems exist, even if monitors don’t directly declare the election fraudulent.2
Russia is a clear example of a country where persistent international monitoring has floundered. The Kremlin has too much power, even over the organizations that monitor it. Thus, another flawed election recently transpired, which monitors could do precious little about. The preliminary OSCE election observer report did point to all the glaring problems and called the election "clearly skewed in favor of one candidate." But that was about all the observers could do. The domestic observers fared little better, their figures ignored by the powers that be.
Countries plagued by violence such as Guyana or Kenya also make it difficult for monitors to promote reforms without fundamental shifts in the underlying conditions. When political victors control not only government policies but also windfalls in jobs and opportunities for their supporters, as in Lesotho for example, it is even harder to get incumbents to adopt recommendations that may lessen their chance of keeping power. And sometimes daunting logistical issues in vast and poor countries like Indonesia make it very hard to implement meaningful electoral reforms. Yet these intransigent situations are what international monitors face time and time again.
The Risks of Election Monitoring
If international election monitoring simply achieved mixed results, the road ahead would be clear: devote more resources to monitoring and press full steam ahead. Unfortunately, it’s not that simple. International election monitoring can be problematic if monitors inadvertently bolster problematic regimes or decrease citizens’ confidence in the sincerity of the international community’s efforts.
The risks are exacerbated by the fact that the quality of monitoring organizations varies considerably. As the mix of organizations has grown more diverse, the field has become very crowded. This fosters unhealthy competition for attention and influence. Furthermore, some groups such as the Commonwealth of Independent States, a post-Soviet club, exist to deliberately counter more credible and critical monitoring organizations, usually from the West. Countries like Russia in particular have figured out how to play the game by constraining critical observer groups while making sure to invite friendly ones to counteract expected criticism.
Yet the issue is not simply a few bad apples that are easy to identify and dismiss. Monitoring activities are inherently political and even the most experienced and credible monitoring organizations worry about funding, pleasing important parties, not stoking violence, and the viability of their ongoing projects in a country. When elections constitute major progress but significant problems and instability remain, as in Cambodia in 1998 or Kenya in 1992, monitors struggle with whether to call the glass half full or half empty. In Kenya in 1992, the International Republican Institute (IRI) followed up a critical report with a statement saying, "but from our perspective we feel that this process is a significant step in Kenya’s transition to genuine democracy." Years later, in 2007, the same organization wanted to release a postelection quick count, but ran into trouble with the local U.S. embassy, which did not like the results. Thus, political, financial, and other considerations at times lead monitors to refrain from clear criticism where it is due or even to endorse elections where more caution would have been warranted. This is especially likely in high-profile cases such as Afghanistan when geopolitical stakes are high for those who sponsor the monitors, or in potentially pivotal elections like those coming up in Myanmar.
This is hardly news to election monitoring organizations. Many are seeking to improve by, for example, making better use of technology and streamlining data-gathering processes. Yet because the monitoring organizations lack resources and independence, they cannot monitor themselves or fix all the problems that exist. The sometimes-dangerous nature of the work, as evidenced by Egypt’s recent arrest of National Democratic Institute, IRI, and other nongovernmental staff, also makes it difficult to focus on reforms.
Organizations know that election monitoring works best when domestic conditions are somewhat favorable and countries are already in transition, as in the post-communist countries in the early 1990s, and when organizations can engage consistently and sustain that engagement. However, organizations rarely have a pot of resources that they can simply allocate as they see fit. Instead, they are at the mercy of donors or governments that sometimes want monitors to go to countries such as Afghanistan and Pakistan to be seen, even if they have little hope of being effective. Similarly, a shortage of resources and the need to gain media attention to sustain the organization’s image makes it harder for organizations to focus on proper follow-up after an election is over. Often, it’s on to the next election right away.
In 2005 the United Nations spearheaded a Declaration of Principles for International Election Observation and Code of Conduct for International Election Observers. The ideals embedded are admirable, but for monitoring to change on the ground, monitoring missions will need to gain greater independence so that they do not have to run their assessments by their donors or worry about their long-term survival. The paradox is that such decisions rest with their sponsors, who are guarding their own interests. A sponsor willing to provide complete freedom credibly and consistently across all elections probably does not exist. If it were possible to relocate general election monitoring to a more central UN-like setting, for example, as was indeed attempted in the early days of election monitoring, monitoring might become more neutral, but it would also run the risk of being watered down into impotence. Meanwhile, current dynamics make needed reforms difficult. If we are stuck with the election monitoring system we have, we should continue treating it with both enthusiasm and some healthy skepticism, while making improvements where we can.
Judith Kelley is associate professor of public policy and political science at the Duke University Sanford School of Public Policy. She is the author of Monitoring Democracy: When International Election Observation Works, and Why It Often Fails (Princeton University Press, 2012).
1. Judith G. Kelley, Monitoring Democracy: When International Election Observation Works, and Why It Often Fails (Princeton: Princeton University Press, 2012).