Elections:Lessons from the U.S. Experience in AfghanistanInteractive Summary

To effectively prepare for and administer elections, the Independent Election Commission (IEC)—which is responsible for administering Afghan elections—must have certain qualities, both as an institution and at a staff level, that are in short supply. First, at the institutional level, the IEC must have the necessary structure to adequately address contentious issues. As it is, the laws, regulations, and conventions governing the IEC’s roles, responsibilities, and internal communication hinder decision-making. Second, at the leadership level, the IEC needs individuals with the vision and discipline to plan for long-term success and quickly react to short-term developments. Instead, the IEC’s leadership is often paralyzed by indecision and appears unwilling to take action today in preparation for tomorrow. Third, IEC staff must have the knowledge and skills to carry out their duties. However, post-election staff purges, inexperienced leadership, corrupt hiring practices, inadequate training, and a shortage of qualified job candidates have contributed to a poorly trained and poorly motivated workforce. Fourth, even for the leaders and staff who do have the knowledge and ability to do their jobs, the IEC must have the will and ability to hold them accountable. Yet with a few notable exceptions, IEC personnel have seldom faced consequences for incompetence or fraud, despite the existence of basic legal foundations for accountability.

“It took a huge amount of diplomatic heavy lifting to get the IEC to take an essential decision. This shouldn’t be the role of the international community at this stage of Afghanistan’s democratic development.” International official

UNAMA photo by Fardin Waezi

The creation of Afghanistan’s first national voter registry in 2018 was a major accomplishment and helped reduce ballot stuffing by tying voters to specific polling stations for the first time. However, problems with the registry’s implementation hindered its ability to mitigate fraud. Registering requires a voter to have a national identification card, which is easy to counterfeit, and there is no effective way to prevent or detect efforts to register with fraudulent documents. The number of registered voters is improbably high, given the population size and low turnout shortly after registering, which likely indicates registration fraud.

Malpractice and lack of transparency also undermine the credibility of the voter registry. On multiple occasions, hundreds of thousands of voters were removed from the registry under opaque circumstances. On election days in 2018 and 2019, large numbers of voters arrived at their polling station only to find themselves unable to vote because their names did not appear on the voter lists. To avoid disenfranchising a large number of voters, the Independent Election Commission allowed some voters not on the lists to vote anyway.

DOD photo

Election fraud in Afghanistan is rampant and takes many forms: Political leaders exert influence over senior election officials and, through them, lower-level staff, and election commissioners and their senior staff sell their services for financial gain. Senior election officials thus play an ambiguous role, serving variously as protectors of the process, perpetrators of fraud, illicit collaborators with senior government officials, and victims of their abuses. Fraud is also perpetrated by local powerbrokers trying to curry favor with candidates in the anticipation of reward, in the form of government contracts, jobs, or payoffs. It is difficult to detect and prove fraud, and even harder to reduce it. Anti-fraud measures are often co-opted to perpetrate more fraud, and even successful fraud mitigation can end up suppressing legitimate votes, sometimes in ways that favor one group over another.

“Of all of the places I’ve worked on elections, I have never met a more corrupt or dysfunctional [electoral management body].” International official

UNAMA photo by Fardin Waezi

The transparent resolution of disputes is a critical safeguard for ensuring an election’s credibility. However, Afghanistan does not have a credible dispute resolution process. The Electoral Complaints Commission (ECC)—which is responsible for adjudicating election complaints—is overwhelmed with complaints. Its provincial offices are weak, vulnerable to political influence, and operate with little oversight. ECC officials are unable to make decisions quickly and rarely justify or share them with the public, and referrals for and prosecution of electoral crimes is minimal. Similarly, upon receiving ECC rulings, the IEC rarely justifies or publicizes its own decisions on which votes to ultimately exclude.

These problems are both a cause and an effect of a worrying trend: Election fraud is increasingly centralized in the dispute resolution process at the provincial and central headquarters, where fraudsters can have the biggest impact for the least effort, as well as the fewest witnesses and the thickest smokescreen. Chaos and malpractice in the central and provincial electoral bodies in the resolution of disputes creates ideal conditions for both election commissions to make changes to the results, and since there is no expectation of transparency, perpetrators can commit fraud with impunity. The process that is supposed to rout out fraud is, instead, when some of its most potent forms occur.

“[Organizing vote tabulation] is not rocket science, unless you don’t want to do it. Anytime I see a really chaotic tally center, I am always suspicious that that is not an accident.” Senior elections advisor

UNAMA photo

The 2018 and 2019 elections showed the Afghan government was unable to use technology to improve the credibility of its elections. Despite this, both the government and political parties have sought to continue and even expand the use of technology. Though it did reduce ballot box stuffing, election technology created new vulnerabilities to the transparency and credibility of Afghan elections. In 2018, the Independent Election Commission adopted election technology less than a month before an election, leading to several failures. In 2019, it failed to follow its own procedures for determining how data collected by biometric devices would be used to invalidate ballots, creating opportunities to contest the election. The use of technology in Afghan elections is not inherently problematic, but political and technical challenges are likely to recur if each election continues to feature new, poorly understood, and untested technology.

“Technology brings more confidence where there is already confidence, but it doesn't increase trust in an environment where there is none.” International elections expert

IEC photo

Election observers can increase the transparency and credibility of Afghan elections by publicizing electoral fraud and malpractice. However, their efforts are hindered in several ways. Observers are often intimidated, co-opted, or are themselves corrupt. Insecurity often makes polling centers inaccessible to observers, and even when observers are present, powerbrokers with a stake in the election often coerce them into falsifying reports and ignoring irregularities. Sometimes observers submit false observation reports because the observers are for sale or otherwise unreliable. In addition, election officials rarely share with observers the critical information necessary for them to detect fraud, like which polling stations will be open on election day. Funding for observation organizations and candidate agents is insufficient or comes too late, which leads to poor training. As a result, evidentiary standards in observer reporting are inadequate, often making their reports useless for the electoral dispute resolution process.

IEC photo

Most problems with Afghanistan’s electoral process cannot be blamed on poor donor assistance, but donors make their assistance less effective in several ways. First, to avoid the appearance of meddling in an election, donors often err on the side of caution and miss opportunities to provide proactive solutions to problems—such as advising election commissions on how to find and hire professional civil servants. This forces them to intervene reactively (and sometimes controversially) after an election goes poorly—such as helping broker power-sharing agreements between candidates. Second, because donors are so careful about proactively intervening in elections, they wind up devoting excessive attention to technical problems. Third, the reason donors need to address so many technical problems is because they provide financial assistance so late in the electoral process that technical problems proliferate, requiring troubleshooting. Between elections, donors reduce their electoral assistance, waiting until the final months before an election to ramp up support. At that point, it is too late to build electoral institutions with the capacity to credibly manage an election. The current donor assistance model is effective enough to ensure that elections happen, but not enough to address recurring problems that end up calling the credibility of elections into question.

“Every time these elections have taken place, there is a massive influx of financial support. There is a last minute scramble…and then a couple of years later the lights go out and everyone goes home. You lose the experience, and equipment and materials disappear into thin air.” Dominic Grant, ECC Advisor

Donors focusing on the short-term goal of simply holding an election can miss opportunities to build the capacity of the election institutions and processes needed to sustain a credible election process once donor support ends. Donor support for elections tends to align with the rhythm of each country’s election cycle: The closer the election is on the horizon, the more donors spend supporting it. When the election is over, donors understandably want to shift priorities to address areas that had been neglected while elections took up so much of their time.

However, disengaging from electoral support, particularly in Afghanistan, can affect the long-term development of the election process and the electoral management bodies that implement and oversee that process. Elections require years of work by numerous government agencies, civil society organizations, and political parties conducting simultaneous backwards planning and responding to unpredictable events in concert. The first half of most Afghan electoral cycles is often the time when such efforts would be most helpful. During this time, donors can support Afghan efforts to draft strategic and operational plans, reform the voter registry, hire and train competent staff, develop a framework for using election technology, create an effective and transparent dispute resolution process, and ensure that the electoral process is monitored by independent and competent election observers.

Electoral fraud mitigation in fragile and conflict-affected states is a task that never ends. If fraud is mitigated in one part of the process, it tends to pop up again in another part. Even anti-fraud measures themselves can be hijacked to create a new kind of fraud, and strengthening one part of the process is of limited value when other parts remain vulnerable to fraud.

Anti-fraud measures can help, but they can also be unhelpful or even counterproductive when they are poorly implemented. Malpractice in implementing anti-fraud measures can mask or even enable fraud. Malpractice can also have the same impact as fraud by disenfranchising voters and delegitimizing the election, making it appear rigged even if progress is being made to bolster its integrity.

Even well-conceived and properly rolled out anti-fraud measures will not prevent fraudsters from finding weak links in the electoral process. Such measures can only reduce the prevalence of fraud, and only gradually, as more links in the chain are strengthened. To prevent fraud entirely, fraudsters must be motivated to stop. That motivation is possible with a combination of accountability for electoral crimes and increased faith that deterrence applies equally to one’s opponents.

A lack of transparency not only serves to conceal fraud in the electoral process, it can enable fraud on a much larger scale. Withholding information about how electoral bodies address complaints or exclude votes on a case-by-case basis allows corrupt or coerced officials to commit fraud without leaving a trace. Until transparency in the adjudication process is given as much attention as fraud mitigation at polling centers, election results risk being perceived as illegitimate, even if fraud is actually decreasing.

Transparency is not an election attribute created in the weeks after election day; it is created over years, as electoral bodies create step-by-step procedures for evaluating complaints and discrepancies, share these processes with the public, and demonstrate that the steps were followed after election day by making detailed justifications of all complaints and auditable election data available to the public.

In an election, particularly one with minimal transparency, poorly implemented technologies only create more questions about the legitimacy of election results. As one international official observed, election technology “brings more confidence where there is already confidence, but it doesn’t increase trust in an environment where there is none.” In the 2018 and 2019 elections, weak and hasty planning undermined the use of technology: Instead of reducing fraud, election technology became the new basis upon which to contest the election. Yet when technology failed to protect the integrity of the 2018 election, the solution was simply to add more technology, not more planning.

The public’s perception of an election’s credibility can determine the extent to which a government is perceived to be legitimate. The officials who manage, oversee, and adjudicate elections have a direct impact on election credibility. Those officials should be particularly skilled at communication, planning, and program management. Election officials also require a combination of technical expertise and a commitment to transparency that can be hard to find in many countries.

Unfortunately, many election officials are not reliable stewards of the process or advocates for reform; they are the problem. The same election officials entrusted to address corruption in the electoral process may use their positions to commit fraud. Donors should not focus all their efforts on the short-term goals of just holding an election, but on the more fundamental and long-term goal of building credible election institutions staffed by competent and trustworthy officials.

Helping Afghans build their long-term capacity and mature their election processes is more likely to happen in the first half of the five-year election cycle, when neither Afghans nor donors are scrambling to ensure an election is held. This would require donors to maintain continuous engagement with the Afghan government throughout an election cycle rather than deprioritizing support after an election to focus on other issues. This can be a challenge. The heavy workload and time required to support Afghan elections can lead to donor fatigue once those elections are over. The high turnover of Afghan election commissioners and their staff exacerbates this fatigue.

One way to ensure that donors maintain continuous engagement with the Afghan government is through the current Electoral Support Group. The Electoral Support Group consists of donor country ambassadors and their staff, and meets in Kabul to discuss political and technical challenges related to elections. This forum allows donors to exchange ideas and speak with a unified voice to the Afghan government, as they did in 2018 when they refused to purchase biometric election technology six weeks before that year’s election, correctly anticipating the operational issues that would occur if such technology were used.

The Electoral Support Group’s ability to help the Afghan government improve their electoral process is limited by the ebb and flow of interest following an election: The group meets weekly during the final year before an election but rarely earlier in the cycle, when Afghan counterparts may be struggling to implement electoral reforms and building capacity at the election commissions. SIGAR’s analysis suggests Afghan officials benefit from continuous engagement with donors to hit their election milestones and fulfill their legal obligations. The Electoral Support Group should be maintained continuously, even if only at the working level and meeting only bi-weekly or monthly. Meeting regularly encourages donors to identify political and technical priorities throughout the five-year electoral cycle, and it would prevent electoral reform and capacity building from being neglected.

In the typical five-year electoral cycle in Afghanistan, donors tend to cram most assistance into the final 18 months, when important preparation for election day is taking place. But just as critical are the previous three and a half years, when reforms are considered and finalized, the capacity of the election commissions’ staff in Kabul and in the field is built, and relationships between technical advisors and election commission leadership and staff are solidified. USAID should maintain steady support for the election commissions to help avoid the loss of capacity that often follows an election.

While Afghan election commission leadership and senior staff are often fired after each election, this did not happen after the 2019 election, which provides an opportunity to build long-term institutional capacity.

Embedded technical experts are also an important part of providing support to Afghan election commissions as they navigate a complex reform process and the early development of strategic, fraud mitigation, and operational plans. In the two years after the 2014 election, the UN’s electoral assistance program had only four staff; this number ramped up to around 70 for the 2018 election and topped out at nearly 180 for the 2019 election. Instead of going from a skeleton crew during non-election periods and then ramping up staffing in election years, donors would be better served by increasing their non-election-year staffing. While it is normal and appropriate for the number of advisors and support to gradually increase throughout the cycle, that support should be substantial from the beginning, befitting a proper reform campaign.

While Afghanistan is a compelling case study, the uneven distribution of electoral assistance highlighted in Recommendation 2 is also a critical problem for electoral support globally. For two decades, evaluations of electoral assistance in dozens of countries conducted by the United Nations, the European Commission, the UK Department for International Development, the Canadian International Development Agency, the International Foundation for Electoral Systems, the European Centre for Electoral Support, the International Institute for Democracy and Electoral Assistance, and even USAID have described how heavily concentrating electoral support in the final stages of an election cycle significantly limits the effectiveness of that assistance.

As one UK analysis noted, “a consistent lesson from international experience is that a succession of short-term, event-based interventions with little continuity from one election to the next achieves few sustainable results.” For this reason, a 2006 European Commission analysis distinguishes between “election assistance” that helps the host nation prepare for election day itself and “electoral assistance” which builds credible institutions and capacity over the long term. Donors often neglect the latter around the world, so this challenge cannot be addressed in Afghanistan alone; it requires a cultural shift across the donor community for all election assistance globally.

USAID assistance to electoral bodies should not be event-driven but more akin to the long-term capacity- and institution-building provided to normal host nation government ministries. If USAID spreads its electoral support more evenly across a country’s electoral cycle and staffs these efforts early with an eye towards capacity and electoral reforms, its assistance will be more likely to result in sustainable electoral processes.

Election observers act as neutral monitors of the electoral process. They increase voter confidence by publicizing what they observe during voting and counting at polling centers, as well as vote aggregation, recounts, and audits at the regional and national level. Their reports also help electoral bodies determine when certain votes must be invalidated. Party and candidate agents also deter and publicize fraud and malpractice, but without the pretense of neutrality. In Afghanistan, USAID and other donors have supported domestic observation organizations, which need funds to train and deploy their observers around the country. In contrast, donors have supported party and candidate agents through training only.

However, support for electoral observers and agents consistently comes too late. It was only weeks before the 2018 parliamentary election when Afghan civil society groups began receiving USAID funding to train and deploy observers and to train candidate agents. The funding came so late that USAID implementing partners had to resort to triage: Candidate agent training was deferred to the 2019 presidential election, and domestic observer groups had insufficient time to train observers on the ethical, substantive, and logistical challenges of being observers in polling stations across the country. The 2019 election was better, but only because it came after the funding mechanisms were already in place.

Given that late funding is a recurring issue, an electoral observation program should be designed 18 months before election day. That way, six months before election day, funds can be disbursed and observation organizations can be ready to monitor voter registration drives, candidate vetting, decisions on which polling centers will open, voting on election day, and the adjudication process afterward. As the earlier stages in that six-month period are less costly, funds could be released in tranches, giving observation organizations time to identify problems, develop deployment and staffing plans, draft and print training manuals and checklists, and properly recruit and train short-term observers.

Voters cannot vote if they cannot both register and access polling stations. In Afghanistan, these two aspects of elections are at a risk of election rigging, which can start well before election day. The national voter registry and the voter registration process are exceptionally vulnerable to manipulation and mismanagement that undermine the registry’s purpose of ensuring credible elections.

The new voter registry was an important electoral reform and has the potential to mitigate substantial risks of fraud. However, the registry is not yet reliable. First, it needs quality control. With so many voters unable to find their names on voter lists in 2018 and 2019, USAID should support IEC efforts to ensure updated voter lists are properly displayed at polling stations across the country. Doing so multiple times over the course of the coming electoral cycle will give voters ample opportunity to make objections and see corrected lists at their polling centers long before election day.

Second, the registry needs anti-fraud measures. Its quality depends on the integrity of the national identification documents (tazkeras) that populate it. Using Ministry of Finance funding, the National Statistics and Information Authority is currently building a database of all 22 million paper tazkeras against which the voter registry can be automatically compared, so that voters who registered with fraudulent tazkeras can be detected and removed from the registry. Being able to make such comparisons seamlessly is a critical safeguard for the registry, so building this tazkera database properly and making it compatible with the registry would benefit from USAID’s attention and support.

Election technology has not had a significant impact on the credibility of Afghan elections, but has merely added another means of contesting them. The 2018 and 2019 elections showed the Afghan government was unable to use technology to improve the credibility of its elections. Despite reducing ballot box stuffing, election technology created new vulnerabilities to the transparency and credibility of Afghan elections.

The United States should support IEC efforts to refine the way it currently uses election technology at polling centers, rather than experimenting with additional technologies. Specifically, the IEC will need to focus on improving the quality of biometric voter verification and electronic transmission of results. If implemented with care and across extended timelines, these two current technologies can help mitigate fraud.

The biometric voter verification process and safeguards associated with the biometric devices have suffered conceptually and operationally due to last-minute rollout, and would greatly benefit from refinements, not innovations. Even before the 2019 election, some election stakeholders were keen to expand the use of biometrics to include a national biometric voter registry, which would commit donors to a new and costly nationwide biometric registration drive before every election. If donors have an appetite for such a long-term investment, supporting the rollout of an e-tazkera (electronic national identification card) would be more sustainable than a stand-alone biometric voter registration, and would serve the same purpose. Experts estimate that e-tazkera would likely not be ready for integration into the electoral process for at least 10 years. However, e-tazkera would have the additional advantage of granting and expanding access to various services like any national civil registry, including education, health, property ownership, justice, and credit.

Donors, Afghan election commissions, parties, and election observers should finalize a concept note at least two years before every election outlining exactly what election technology will do in the coming election and how it will do it, allowing sufficient time for operational planning, procurement, testing, and training.

Electoral dispute resolution is the adjudication of election-related disputes, complaints, objections, or alleged violations of the law, the most serious of which can challenge election outcomes. Effective electoral dispute resolution lends credibility to an election by serving as a peaceful means for contesting election results. In Afghanistan, the Electoral Complaints Commission is charged with resolving electoral disputes.

In the division of technical assistance resources, support to the ECC is far less than the IEC’s share. Furthermore, the ECC’s work has been especially hurt by the uneven nature of electoral assistance, highlighted in Recommendation 2. For example, the ECC received virtually no assistance from 2015 to 2017, leading to a critical loss of capacity and oversight. By 2018, the ECC was particularly unprepared for the parliamentary election and overwhelmed by the volume of complaints. It used spreadsheets to track more than 20,000 complaints, posted very few rulings on its website, and provided almost no legal justification for its decisions to the complainants or the public. Technical solutions, including a complaints database, were created with donor support in 2014, abandoned when support dried up, and then re-implemented at great cost in 2019. Even in 2019, spreadsheets were used yet again because provincial ECC officials lacked the capacity and even email addresses to use the new database.

The insufficient support manifests in personnel allocations as well. During the 2019 election, there were 87 UN advisors embedded at the electoral bodies—42 international and 45 Afghan. Of these, only eight international advisors and two Afghan advisors were embedded at the ECC. The rest were assigned to the IEC.

Despite the relative neglect of the ECC, SIGAR found that the most harmful fraud is migrating from the polling center level to the more centralized dispute resolution process led by the ECC. Therefore, the balance between IEC and ECC technical assistance would benefit from shifting somewhat toward the ECC to ensure the adjudication process is more transparent and credible at a time when the ECC’s work is increasingly vulnerable.