In this post, HKS student Juliette Keeley, MPP ‘17 delves into the challenges of election monitoring and highlights innovations designed to address different aspects of this complex problem. She lays out the advantages and shortcomings of using technology in various capacities to improve election-reporting mechanisms, to report and limit violence and intimidation, and to map community-based organizations. Keeley finds that widespread on-the-ground mobilization and citizen participation are common themes running throughout the most successful election monitoring initiatives.
By Juliette Keeley
Election monitoring is a messy business. Thousands of observers are deployed across a country in the attempt to observe voting. They must quickly report incidents and results to a central unit, in the hopes of generating a credible report on the elections. All this should happen prior to the government’s release of official results. However, reporting is plagued by poor infrastructure and communication systems and, in some cases, violence and intimidation.
In this volatile context, a number of strategies are emerging to improve election monitoring: using text messaging to quickly report election irregularities results, engaging citizens through crowd-sourcing to report violence, and mapping community-based organizations to strengthen democratic institutions.
A common thread runs through these innovative strategies: the push to involve a greater number of on-the-ground, local actors in improving the quality of elections.
Defending democracy through observation
Election monitoring in emerging democracies has grown extensively over the past 20 years. Election monitoring organizations include international governmental organizations, such as the United Nations; non-profit monitoring organizations, such as the National Democratic Institute and the Carter Center; regional organizations like the African Union; and local grassroots organizations.
Regardless of size, they all operate on the premise that the right to vote is only meaningful if elections are credible. Election monitoring organizations aim to defend the right to vote by disseminating information about electoral processes, by increasing transparency, and by building public confidence around elections. They don’t prevent fraud, but rather monitor and report in order to hold politicians accountable and promote
Monitoring across time and space
To achieve these goals, monitoring bodies operate prior to, during, and immediately after elections. Their trained, accredited observers are dispatched in polling stations across a country in order to provide a complete and credible assessment of elections. They assess whether eligible voters were able to freely cast ballots, whether poll workers followed appropriate procedures, whether any sort of disruption occurred, and crucially, whether the results of the ballot count reflected the will of voters.
One major challenge is that information must be gathered and analyzed quickly in order to provide a credible assessment before the release of official results. By independently tabulating results, monitoring bodies can project election results prior to the release of official results, which can deter a government from releasing false reports.
Unfortunately, this has not always been the case: in 2013, the European Union failed to provide a timely report on elections in Kenya, which the International Crisis Group called fraudulent. These geographic and time constraints mean that monitoring bodies need a rapid reporting system that can move large volume of info quickly and reliably. A number of creative applications of existing technologies show promise in addressing these and other challenges to effective election monitoring.
Using biometric voting and SMS reporting: silver bullets?
One reliable and easy way of informing and engaging citizens is through the power of Information and Communication Technologies (ICTs). If well integrated, ICT can yield massive improvements in counting expediency, voting, fraud prevention, and cost reduction, amongst others.
Biometric voting technology is on the rise in developing countries, particularly in African countries. With this technology, voters receive a permanent voter card, which stores biometric information like fingerprints and a picture ID. This voter card is checked at the polling station by a voter card reader, which verifies the voter’s name on his or her card. Ghana, the DRC, Togo, Guinea Conakry, Uganda, Angola, Nigeria, and Mozambique have all begun to use these technologies.
This technology is often costly. While in established democracies, elections cost $1 to $3 per voter, biometrics coupled with poor infrastructure have resulted in skyrocketing elections costs. For example, Kenya’s 2013 election cost a staggering $293 million, with donors such as Canada, the EU, German, the UK and the US contributing $100 million. The polls cost $20 per voter.
Biometric technologies may reduce vote rigging, but they do not constitute a “democratic panacea.” Although the 2013 elections in Sierra Leone utilized biometric technologies, they were impeded by thugs who were sent around to scare voters away from polling stations. Still, if the technologies prove to be accurate, efficient, and sustainable, they can build credibility because
Mobile phones have been used to improve election monitoring, to report on human rights abuses, to strengthen civil society, and to democratize the flow of information. In election monitoring, text messaging, or SMS technology, has the advantage of moving information instantaneously from a text message to a database without human intervention and at low cost.
The National Democratic Institute (NDI), a nonprofit organization that supports democratic development worldwide, has in recent years supported nonpartisan domestic monitoring organizations in using text messaging. In 2005, in Indonesia, observers used text messaging to connect to a call center, which managed as many as 750 observers in four rounds of local elections. Volunteers no longer had to pay for the cost of the call.
In the 2006 elections in Palestine, NDI combined SMS technology and digital mapping using Geographic Information System (GIS) software. In a pilot program, two observer teams text-messaged their polling station code to a cell-phone connected to a laptop, which then downloaded, processed and stored the messages. GIS mapping software transformed the database into a live map of observers’ locations. The NDI, in partnership with the Carter Center, reported the elections to be “a professional and impartial performance.”
In Albania, 2007, a coalition of eight organizations oversaw 1,200 observers across the country in almost 3,000 of the 5,000 polling centers. SMS was used to report turnout and incidents via more than 41,000 text messages that were processed through an international SMS gateway service provider. Observers reported 168 critical incidents of violence, intimidation or harassment that were followed up on by the coalition’s call center the following day. Incidents SMS reports were used to evaluate monitoring organizations’ quality through the reporting of their observers.
SMS technology is not without issues—for example, it can be and has been threatened by blockages and outages. In Albania, one of the mobile phone providers blocked traffic to and from the hotline during testing, and the CDT had to negotiate to restore service. Still, SMS technology allows for quick reporting to the public, which increases participation and trust in elections.
Early warning dashboards: predicting violence in elections
Researcher Clément Vergne finds that violence, intimidation and fraud have been the leading cause of low voter turnout in most developing countries. Vergne finds that the African continent has the lowest voter turnout in the world, at 65%. Low voter turnout has been registered in countries such as Uganda, Nigeria, and Zimbabwe, according to researchers Collier and Vicente. The issue of low voter turnout in many developing countries has ignored.
Some have addressed these issues by involving citizens directly in the monitoring process. Ushahidi is a platform that was developed to map reports of violence in Kenya after post-election violence in 2008, and has been used in several countries to monitor election-related violence. It was used in Ethiopia in 2010, where citizens could directly report on violence, which would then be mapped in real time. The system only received 15 reports and was visited 62 times on the day of election, suggesting that not enough people knew about or used the system to have an impact.
In Nigeria in 2015, Ushahidi worked with the Stakeholder Democracy Network (SDN) to create an early warning dashboard for violence before, during and after elections, with the ultimate goal of trying to predict violence. Using SDN’s extensive networks and 10-year experience on the ground, Ushahidi collected more than 100,000 documents about the Nigerian election, through Twitter accounts and five mainstream media sources. Users could report by sending a text message, tweet, email or by using an app.
During the Armenian elections of 2012, civic activists from Media Diversity Institute and Transparency International Armenia launched iDitord (“iObserver” in Armenian) to monitor the election process. Again, the Ushahidi platform was chosen because it allowed contributors to send reports through multiple means (website, mobile apps, SMS and twitter using the #iditord hashtag) and to create a real-time map.
In order to avoid reports coming from only a small subset of the population, a network of NGOs, political parties and internet activists launched a public awareness campaign. More than 1000 reports were submitted between April and May. Verification of reports was also an issue: as reports of bribery came in, the police and the Central Electoral Commission claimed some reports were erroneous. A rule was created to avoid this issue: reports including by video or photo were marked trustworthy only after being vetted by established journalists or observing NGOs. In this way, 887 out of 1141 reports were marked trustworthy.
Cyber security proved to be crucial: in the months leading up to election, reports came in at about 15/day. On election day, iDitord received more than 5,000 unique visitors and about 500 reports. The platform then suffered a cyber-attack that shut it down for 4 hours.
Crowdsourcing election monitoring is contingent upon citizens’ participation and knowledge of the platform. Civic activists in Armenia had to launch a public awareness campaign to get enough users. Regardless of whether users are able to report election irregularities, they will need Internet access to view the map itself. Coordination between civil society actors is key not just to launch a campaign, but also to verify reports and to publish trustworthy data. Finally, security is a crucial issue because these high-profile sites are likely to be targets of cyber-attacks orchestrated by governments or groups that profit from manipulating election results and are threatened by election monitoring.
Digital technologies are not useful without users—and the key to participation is an infrastructure of local organizations with the capacity for successful engagement.
Grassroots democracy: funding community-based organizations
Consider the case of the Democratic Republic Congo. The country is poised for national elections in November 2016, potentially marking its first peaceful transfer of power in history. Joseph Kabila, the current president, has unsuccessfully tried to change the electoral law and constitution to remove the term limit clause. Kabila, in power since 2001, now appears determined to prolong his stay in power by creating a series of administrative blockages that delay the elections.
The International Crisis Group reported that foreign donors are now hesitant to invest funds in the upcoming elections: as of October 2015, the Independent National Electoral Commission (CENI) had received only 17% of funds earmarked for elections. Similarly, investment in civil society actors working on electoral issues has been limited, as donors leverage their resources to maximize their political leverage over the government in Kinshasa. Limited coordination among bilateral and multilateral donors remains a key challenge.
This delay in funding creates a serious problem for domestic observers, who often lack financial resources and technical capacity. Researcher Judith Kelley of Duke University finds that domestic missions often occur right before elections. In the scramble leading up to elections, civil society groups may even enter into competition with each other for scarce resources. These factors lower domestic groups’ ability for long-term planning and effectiveness.
Master in Public Policy candidate Tom O’Bryan at the Kennedy School, who spent two years in the DRC, believes that Congolese community-based organizations there are notoriously underfunded, despite their comparative mastery of local political dynamics.
The problem O’Bryan sees is a lack of information. With limited access to information on potential CBO partners for elections-focused projects, international donors including USAID, the UK Department for International Development and the European Union often regard such partnerships as “risky.” This information gap, combined with the challenge of limited coordination among donors on which projects they support and where, has resulted in funding being disproportionately allocated to local organizations in urban centers.
This concentration is particularly the case for Kinshasa, Goma and Bukavu. Great swathes of the country’s expansive interior are neglected, and CBOs working on electoral issues there receive comparatively limited financial and technical support from international donors.
A secondary consequence of limited coordination among international donors is that some projects and thematic issues remain comparatively under-funded. Investors, unaware of the scope and focus of fellow donors’ support, consequently tend to fund similar thematic projects in similar geographies: civic education projects in North Kivu, for example. Limited access to information on the projects already supported by other donors hinders the development of a robust, well-coordinated international support strategy for democratic elections in the DRC.
To increase donors’ access to information on potential CBO partners, Ben Affleck’s nonprofit, the Eastern Congo Initiative, partnered with USAID to produce an online database of these organizations. Researchers produced a landscape analysis of almost 300 CBOs working on social and economic issues in four eastern provinces. The database included detailed information on these organizations’ thematic priorities, an assessment of their technical capacity, organizational budget, and contact information.
Perhaps unsurprisingly, the initiative found that urban areas and war zones received the most assistance from funders, while rural and comparatively stable regions were neglected. Many capable, small NGOs from these geographies have difficulty accessing funding, in part because of their rural location and the lack of basic infrastructure connecting them to larger cities.
Donors are oftentimes unaware of the very existence of these organizations. With limited access to basic information about these NGOs, funders remain hesitant to make a “risky” investment and provide direct financial support.
Inspired by USAID’s work, O’Bryan is now building, with the support of the Ash Center and other organizations, a series of interactive online maps of the DRC. The first will provide such information on local NGOs working on elections and democratic governance issues across all 26 of the Congo’s provinces. This map will help international donors identify promising local partners from regions of the country that have received comparatively limited technical and financial support.
The second map will visualize all elections-focused projects currently being funded in the DRC. It will divide projects into more than ten categories—from election monitoring to candidate training, from voter registration to elections polling—and highlight geographic and thematic gaps. The second map could help to foster a coordinated investment strategy in Congolese democracy at the grassroots level, minimizing the aforementioned geographic and thematic gaps.
Why fund local partners anyway?
The literature on the benefits and challenges of funding local capacity is quite extensive. International observers from international governmental organizations (IGOs) come with baggage. Election monitoring is expensive, and foreign observers are especially costly, at an average of $600 a day, says Tim Meisburger, the Regional Director for Elections and Political Processes at the Asia Foundation. At $10 a day, domestic observers’ typical wages pale in comparison.
Bound by scarce resources, international representatives are fewer in number and, therefore, largely symbolic. Theoretically, they cover rural and urban areas alike, in randomly selected polling stations through a method known as Parallel Vote Tabulations (PVT).
In reality, the patterns of visits are often fairly predictable because of logistics or security concerns. As a result, there are areas where election observers are rarely deployed, like in conflict-riddled countries such as Iraq or Afghanistan, or in areas poor infrastructure, such as Nepal or the Democratic Republic of Congo. International monitors have been targeted by terrorist groups: during the 2014 presidential elections in Afghanistan, the National Democratic Institute pulled its observers from the country after a Taliban attack on a Kabul hotel that hosted observers.
International monitors also come from organizations that vary greatly in approach and subsequently in their results. For example, competing IGOs monitored the 2011 presidential elections in the DRC. While the Carter Center, the National Democratic Institute and the European Union declared the voting fraudulent, African missions such as the African Union and four others declared the polls successful. Kelley finds that regional IGOs with more non-democratic members are less neutral, as is the case for the African Union. However, any organization tied to national governing bodies will likely have political interests in certain countries that can interfere with the observation mission’s freedom.
On the issue of domestic observers, a greater number can be deployed. Local partners know communication channels to reach citizens. They speak the language, know what kind of irregularities to look for, and know how to publicize their findings, says doctoral candidate at the University of Oxford Lynge-Mangueira.
Geography matters too. The Eastern Congo Initiative showed that there is a wealth of local organizations spread across the DRC. Rural organizations can access hard-to-reach polling stations and provide a more valid assessment of elections than Kinshasa organizations.
Still, funding local monitors is no silver bullet either. By virtue of local ties, credibility and neutrality can be harder to establish. Considering security, domestic observers will stay in the country after elections and in so doing could face retributions. Finally, larger IGOs may be better equipped and more politically connected to high-level government officials, and therefore, they may have more political capital to exert than would small NGOs.
“Elections are a process, not an event”
Clearly, coordination between local and international bodies is key. But when should this coordination take place? Innovations like biotechnology, SMS reporting and early warning systems focus on the moments immediately prior to and during elections. But ultimately, election monitoring takes time and preparation, and funding in places like DRC will be required now through elections in November.
focused on procedures during the polling day, even though the extended period of time leading up to elections offers many opportunities for abuse. This can include manipulation of electoral rules, media dominance, vote-buying and electoral fraud. Leaders and supporters of the opposition are often targets of state violence.
Leading up to the 2011 elections in the DRC, the Carter Center reported that several serious incidents of intimidation and violence had already occurred.
To curb violence, Dr. Amanda Sives of the University of Liverpool recommends that IGOs maintain a long-term presence during the entire electoral process. Long-term observers can add credibility to the election and develop long-term partnerships with community-based organizations. Lynge- Mangueira and Sives both advise that international observer groups invest more directly in the local capacity building of domestic observers.
Ultimately, grassroots organizations can help bolster local elections in the DRC, which may be the key to sustainable democratic efforts in a country with mired presidential politics.
Election monitoring: assessing symptoms of democracy
Election monitoring is but one of the many tools available to improve the quality of elections; however, for elections to fulfill their intended purpose, they must be bolstered by the people they seek to serve.
This post explores new strategies to improve election monitoring by engaging networks of observers through SMS reporting, networks of everyday citizens through violence mapping, and finally networks of community organizations through targeted funding. None of these strategies alone will guarantee free and fair elections, but they can bring citizens and grassroots efforts a greater voice in the electoral process.
Juliette Keely is a MPP ’17 candidate at Harvard Kennedy School and research assistant at the Ash Center for Democratic Governance and Innovation. Keeley recently worked on a team of researchers in a Harvard University field study exploring the lessons that Europe can learn from Morocco’s experience in, and novel approach to, managing migration.
Collier, P. & Vicente, C. P. (2010). “Violence, Bribery and Fraud: the political economy of elections in Sub-Saharan Africa (Draft).” http://users.ox.ac.uk/~econpco/research/pdfs/ViolenceBriberyandFraudSub-Saharan.pdf
Joint Committee for Monitoring 2006 Elections. (2006). “Monitoring the second round of the Bahrain 2006 parliamentary and municipal elections [Press statement].”
Kalil, T. (2009). “Harnessing the Mobile Revolution.” The New Policy Institute. http://www.mitpressjournals.org/doi/pdf/10.1162/itgg.2009.4.1.9
Kelley, J. (2012). Monitoring Democracy: When International Election Observation Works, and Why It Often Fails. Princeton University Press.
Lynge-Mangueira, H. (2012). “Why ‘Professionalizing’ International Election Observation Might Not be Enough to Ensure Effective Election Observation.” International Institute for Democracy and Electoral Assistance. http://www.idea.int/democracydialog/upload/why-professionalizing-international-election-observation-might-not-be-enough-to-ensure-effective-election-observation.pdf
Riera, A. & Brown, P. (2003). “Bringing Confidence to Electronic Voting.” Scytl Online World Security.
Schuler, I. (2008). “SMS as a Tool in Election Observation.” National Democratic Institute. http://www.mitpressjournals.org.ezp-prod1.hul.harvard.edu/doi/pdf/10.1162/itgg.2008.3.2.143
Stein, M. (2006). “MobileActive strategy guide #1: Using mobile phones in electoral and voter registration campaigns.” MobileActive. https://www.ndi.org/files/MobileActiveGuide1_English.pdf
Vergne, C. (2009). “Turnout in Developing Countries: The Effect of Mass Media on National Voter Participation.” HAL: Archives-Ouvertes.