5. What We Know About Disputed Elections

At a minimum, electoral integrity can be understood to be based on two key principles: the credibility of the entire electoral cycle and the legitimacy of a set of global norms.

Given the wide variation in views regarding what is core to a credible election (Elklit and Svensson 1997; van Ham 2015), there is unsurprisingly no consensus on how to define electoral integrity. At a minimum, however, electoral integrity can be understood to be based on two key principles: the credibility of the entire electoral cycle (as opposed to merely voting and vote counting) and the legitimacy of a set of global norms (as articulated through international treaties such as the Universal Declaration of Human Rights, the International Covenant on Civil and Political Rights and others). Norris (2013: 564) articulates the concept as ‘international conventions and global norms, applying universally to all countries worldwide throughout the electoral cycle, including during the pre-electoral period, the campaign, and on polling day, and its aftermath’. Practitioners, through the Global Commission on Elections, Democracy and Security, consider political equality to be an additional core component of this concept (2012: 6). Even here, of course, it may be contentious to claim that ‘global norms’ are ‘owned’ by everyone or are universally applicable (Mutua 2000).

Additionally, the forthcoming Model Commitments for Advancing Genuine and Credible Elections, authored by International IDEA, the Carter Center, the National Democratic Institute and the International Foundation for Electoral Systems, flesh out and operationalize these norms and guide countries in identifying where obstacles to holding fully democratic elections might lie. These commitments are tied to the international norms described above and to standards and best practices, as expressed in regional treaties and national constitutions. Together, these commitments help elucidate the global norms that inform the meaning of electoral integrity.

Despite the lack of consensus, there are several reasons why understanding electoral integrity—even broadly defined—is critical to democracy. First, while elections are not sufficient to qualify a political system as democratic, it is well established that they are a core minimum for any democracy. As such, a flawed or farcical election is uniquely detrimental to the legitimacy of a democratic system. Indeed, countries that hold elections without a bare minimum of the enabling rights required for participation are seen as ‘electoral autocracies’ (Lührmann, Tannenberg and Lindberg 2018). Such environments demonstrate that countries can hold elections that do not substantively contribute to democracy. Clearly controlled electoral environments are popularly understood as little more than ‘shams’, evidenced recently by the severely repressed media, the lack of opposition parties and the climate of fear that marked the Belarusian and Russian elections (Komin 2024; Liubakova 2024). Conversely, ‘fairly conducted and regular elections create system legitimacy’ (Banducci and Karp 2003: 443).

Second, a robust concept of electoral integrity is key for reform. Practitioners and policymakers, including EMBs, election support organizations, legislators and courts, and election observation groups, must understand where the weaknesses in the electoral process are so that they can work to strengthen those activities and procedures.

Fortunately, the knowledge base on election integrity is rich, and expert data sets provide an excellent source of information. The Electoral Integrity Project (EIP), for example, which annually surveys experts in 169 countries on a comprehensive set of questions covering 11 aspects of the electoral cycle, provides a long record of expert evaluations of electoral integrity. Since at least 2013, the EIP’s Perceptions of Electoral Integrity (PEI) data set shows that campaign finance and campaign media coverage are the two weakest aspects of electoral processes around the world. Those long-standing concerns now co-exist with new ones about money spent online for campaign purposes. For example, some analyses have illustrated the need to embed digital campaigning rules within the regulatory architecture for political finance (Tham et al. 2022).

Contrary to expectations, EIP data show that vote tabulation and the announcement of results have historically been among the least problematic stages of electoral processes (Norris, Frank and Martínez i Coma 2014; Norris, Martínez i Coma and Grömping 2015; Norris et al. 2016; Norris and Grömping 2019; Garnett et al. 2023a, 2023b). Notably, however, a recent study of the PEI data set (2012–2022) found that vote counting has emerged as an area of significant decline, even though the degree of change is relatively small (James, Matlosa and Shale 2023: 327). These recent findings are important to consider, especially in comparison with the GSoD Indices data. The latter data set shows that metrics of Credible Elections, which include assessments of intentional irregularities and voter fraud, government intimidation, EMB independence and capacity, fairness of the legal framework and political competition, have been among the most broadly declining indicators of democratic performance for at least the last five years (International IDEA 2023i). Specifically, GSoD Indices data show that some of the most common problems include government intimidation or harassment of opposition candidates, intentional irregularities, declining EMB autonomy and a sense that the overall process was not free and fair.

EIP data show that vote tabulation and the announcement of results have historically been among the least problematic stages of electoral processes.

Given the fears around election-related violence, it is somewhat surprising that it does not feature among the list of the most urgent problems for electoral integrity. Experts do consider violence to be important, and one study found that reducing the threat of violence at the polls is indeed likely to have an impact on assessments of electoral integrity (Frank and Martínez i Coma 2017). Most other studies on violence, however, focus on how it impacts voters’ behaviour (as opposed to overall integrity), which is taken up below.

Somewhat in contrast to this, election observers often focus on violence (or the lack thereof) in their official reports (Matlosa 2021). Here, there are no comprehensive studies that link observer reports and expert assessments of electoral integrity. Judith Kelley’s (2012) seminal work has shown, however, that there are cases in which observer recommendations do eventually contribute to improvements. She demonstrates that such cases tend to be those in which there is a desire for increased cooperation with the West, where there is domestic pressure for reform and where observers are invested in long-term follow-up and consistent engagement in and with countries. Observers’ ability to deter fraud in some cases is also shown through case studies (Asunka et al. 2019). Negative statements in election observer reports are an additional sign of flawed electoral processes, at both the local and international levels (Kelley 2012). However, in contrast to the positive effects of established international observation practices, there is growing concern about fake electoral observation missions. These are used to advance the interests of certain political actors and do not adhere to the common standards and good practice of international election observation, as defined in key documents such as the Declaration of Principles for International Election Observation, adopted in 2005 (National Democratic Institute 2005). Examples of biased missions include the 2023 local and regional elections in occupied territories of Ukraine, the 2022 referendum in occupied territories of Ukraine and the 2021 parliamentary elections in Crimea (European Platform for Democratic Elections n.d.). An in-depth study of observation missions, however, is outside the scope of the present study.

In recent years, scholars have broadened the scope of their work to include popular perceptions of electoral integrity, responding to a serious vacuum in the knowledge base (Pearce Laanela 2023). Understanding how the public perceives and considers elections is important for several reasons. First, low confidence in elections can lead to decreased turnout (Birch 2010).

Second, it is voters who ultimately legitimize elections and the resulting elected government; their opinions are thus crucial for political stability, the popular endorsement of a legitimate set of elected leaders and the establishment of the resulting institutions of democratic governance. McAllister and White (2011: 665) explain, ‘The concept of “fairness” is a qualitative judgment that voters themselves are in an ideal position to make … . It is voters who experience the election campaign, directly through contact with election candidates and parties, and indirectly through the mass media. Unlike outside observers, they cast a ballot and therefore have a degree of “ownership” over the election process and its outcome’ and a direct perception of how credibly, freely and fairly that process has unfolded. People’s beliefs in and support for certain platforms and parties give them a personal stake in outcomes and processes. In fact, scholars have made significant advances in understanding how partisanship contributes to popular evaluations, with clear findings that a ‘winner’s effect’ plays a strong role in popular perceptions of fairness, as voters’ assessments of elections are tinged by the success (or failure) of the party they supported (Alvarez, Hall and Llewellyn 2008; Ansolabehere and Persily 2008; Wilson and Brewer 2013; Bowler et al. 2015; Sances and Stewart 2015).

Third, elections are pivotal to citizens’ trust in their country’s broader political system. When there are doubts about electoral integrity, people (and especially supporters of the losing party) are more likely to doubt the fairness of an election and the legitimacy of its outcome, which results in lower levels of trust (Mauk 2022). In unfair electoral contexts, voters’ mistrust extends to political parties, the legislature and government (McAllister and White 2011). Indeed, the potential of cracks in perceptions of electoral integrity to trigger a domino effect which leads to doubts about the legitimacy of the broader democratic model of governance is perhaps the most important reason for focused attention on the drivers of electoral integrity.

Recent evidence suggests cause for concern. International IDEA’s Perceptions of Democracy Survey showed that in 11 of the 19 countries surveyed, less than half of the respondents said that the most recent elections in their countries were free and fair (see Figure 5.1). In some cases, the percentage of the respondents who agreed with expert assessments was quite low (see Figure 5.2). Marginalized communities, including self-identified minorities and those within the lowest income groups, were more sceptical than others (see Figure 5.3) (International IDEA 2024g). The survey results echo others’ work, including a study in the USA which found that women and minorities were less likely to think that elections were fair. Women of colour were ‘much less likely than others to see officials and vote counts as fair’ (Bowler et al. 2015: 6).

Scholarship demonstrates that public perceptions about elections may be driven by a wide range of factors, some of which are clearly outside the realm of election administration. A recent survey in the context of the USA showed that people with lower levels of education and who are younger are more likely to mistrust elections. Those who earn more are more likely to trust elections (Kousser 2023). In the US context, there is also evidence that white Americans who believe they (and others in this racial category) are the victims of discrimination are less likely to trust elections (Filindra, Kaplan and Manning 2024). In the context of Brazil, a 2023 study found that right-wing ideology and exposure to voter fraud allegations are correlated with higher levels of mistrust, raising concerns about the influence of political elites who spread disinformation for their own political ends (Rossini, Mont’Alverne and Kalogeropoulos 2023). In the United Kingdom, a 2021 study found that declining levels of political trust were tied to the perception that politicians were out for themselves rather than committed to the best interest of the country (Quilter-Pinner et al. 2021). Clearly, many of these drivers are outside the ambit of EMBs or other agencies involved in elections.

Some factors, however, are more directly related to the work of those who administer and make decisions about how elections are run. There is evidence that people’s confidence in electoral integrity is tied to the ‘rules of the game’. Birch (2008) demonstrates, for example, that elections utilizing proportional representation systems and public funding options are associated with higher public confidence. In the USA, distrust of elections is driven partly by concerns about postal voting, ineligible voters casting ballots and eligible voters being kept away from the polls because of strict voter identification laws (Kousser 2023). People also pay attention to the practical details of election administration. In their study, Bowler et al. (2015: 8) found that people expressed more confidence about the fairness of an election in US states that had high scores in administrative quality, explaining that ‘technical improvements to electoral administration can improve voter perceptions of elections being fair’. Examples of technical improvements include shorter wait times for voting, the availability of online registration, disability-related voting accommodations and the provision of tools that enable voters to look up election-related information, among other things. Bowler et al.’s findings mirror other work which has found that EMBs contribute to popular confidence in elections (Debrah, Asante and Gyimah-Boadi 2010; Kerr 2013).

Popular perceptions of electoral integrity may also be affected by innovations such as electronic voting, but the impact is dependent on the underlying context. Studies have found that while there are cases in which electronic voting raises specific concerns (e.g. ballot secrecy and manipulation of vote counting or results tabulation), people’s primary concerns related to voting technology are usability, verifiability and the privacy and security of the technology (Alvarez et al. 2013; Avgerou et al. 2007). Trust in electronic systems is itself tied to people’s confidence in the underlying environment. If voters do not trust the authorities responsible for elections or are doubtful about information and communication technology, for example, even the most secure, usable, verifiable and privacy-respecting machines will not be embraced.

Given the technical complexity involved in the adoption of electronic voting, partnerships between EMBs and civil society can be important, as voters may delegate their trust to CSOs and experts to verify systems that are beyond their own capacities to investigate. A good example of this occurred in Switzerland in 2019, when the source code for a new Swiss Post electronic voting system was published, and researchers were able to uncover vulnerabilities, allowing the code to be revised before the vulnerability could be exploited (Swiss Federal Chancellery 2019). This kind of three-party interaction between the voter, an expert group and the EMB is an example of delegated trust (Warren 1999), in which trust in those with the capacity to verify the integrity of the system can act as a guarantor for the EMB. Even where voting takes place through an electronic system, the production of a paper record can reassure voters and facilitate a manual recount in case verification of the electronic record is required (Wolf, Nackerdien and Tuccinardi 2011). In the US context, the use of paper ballots was credited with affirming the accuracy of the 2020 election (Tisler and Baker 2022). In Namibia, the EMB decided to revert to paper ballots in the 2024 elections, in part due to the Supreme Court’s prior ruling that if electronic voting machines were used, a verifiable paper trail would be necessary.

Of course, people also evaluate the credibility of elections based on their own experiences of the electoral process as well as some second-hand information (Kerr 2018). In his study of Nigerian voters’ perceptions of electoral integrity, Kerr (2018: 11) found that negative election day experiences were strongly associated with perceptions of electoral integrity. Specifically, voters who had experiences with ineffective card readers and malfunctioning machines were less willing to express the highest degree of confidence in the integrity of the vote count and less willing to consider the electoral process free and fair. On the other hand, it is important to note that there is evidence that personal experience with fraud or manipulation does not consistently have a significant effect on perceptions of integrity (Shah 2015; Wellman, Hyde and Hall 2018). In some cases, voters may even witness or expect fraud without subsequent impacts on their assessments (Schedler 1999). The role of partisan identity (see below) may explain this finding.

A portion of the second-hand information that Kerr refers to may come from election observers. The assumption is that the public will view elections as flawed if election observers report problems. Hyde and Marinov (2014) assert that this is particularly the case when accurate and credible information is limited. In his work in Nigeria, Kerr (2018) found that voters tend to express greater confidence in national elections when they witness the presence of international observers. When they witness only domestic observers, they are more likely to express confidence in the integrity of elections at their polling stations. Another more recent study, however, found that observer reports can have effects that are in opposition to what they intend (Benstead, Kao and Lust 2020). In some contexts, what is perceived as Western influence on domestic politics is resented, and people adjust their evaluations to be in opposition to observer reports. This study found that positive statements from international observers in Jordan, for example, induced some citizens to make statements about the 2013 elections that were more negative than they otherwise would have made because they were influenced by their pre-existing negative beliefs that foreign governments were intervening in their country. The fact that they did not like the source of the positive assessment of the elections made them take a more negative view of it (Benstead, Kao and Lust 2020).

Finally, it is important to consider the influence of violence on public perceptions. Unfortunately, the scholarly literature on the specific connection between the occurrence of violence and people’s opinions of electoral integrity is limited. One study showed that the use of violence decreased the public’s support for candidates who make use of it, but it did not show a link between violence and overall integrity (Rosenzweig 2021). Another study showed that violence reduces turnout, but the question of perceptions of integrity were not taken up (Bratton 2008). One recent study has made inroads, showing that in polarized contexts where there is violence, people with strong partisan identities tend to assess elections as being more free and fair if co-partisans were involved in the violence (Daxecker and Fjelde 2022). The sample size here is small, and results are thus difficult to generalize beyond this case.

In sum, people’s perceptions of electoral integrity are driven by a wide range of factors. These include their trust in political leaders, exposure to fraud allegations, trust in special voting arrangements and their own (and their peers’) experience at polling stations. The rules also matter, and factors such as public funding options and proportional representation systems are correlated with higher levels of confidence.

The wealth of scholarship cited in this report provides a critical baseline of knowledge about election integrity, but there are now new factors to consider. In addition to long-standing threats to integrity (such as bias in media coverage and weak campaign finance regulations), electoral processes today face a new array of risks, ranging from the practical consequences of severe weather events (International IDEA 2024j) to elaborate and hard-to-mitigate cyberattacks, the unregulated use and misuse of AI technology, disinformation campaigns and the rise of fake electoral observation missions (International IDEA 2024g; European Platform for Democratic Elections n.d.). These are in addition to the broader system-level drivers mentioned above, including declining levels of trust in government, disillusionment with political elites, etc.

International IDEA lists eight categories of potential threats to electoral integrity to reflect this context. These threats are undemocratic electoral reforms, electoral management and dispute resolution malfunctions, electoral malpractices (fraud and corruption), violence, malicious online actions, lack of trust and negative public perceptions (without substantiation), environmental and man-made disasters, and gender-based discrimination and gender-based violence. Depending on the context, some threats may be more or less relevant at various times.

Perhaps most urgently, public faith in electoral integrity today is threatened by false narratives—often fuelled by opportunistic politicians and often without evidence—that seek to discredit elections. Such threats can have dire impacts on public trust. These narratives have been most clearly evident in Brazil and the USA, but they have also impacted countries such as France (Adler and Thakur 2021) and India (Anand 2024).

Public faith in electoral integrity today is threatened by false narratives that seek to discredit elections.

Disinformation in particular causes significant worry, but there is little systematic data to demonstrate the severity of impact on voters, their behaviour and resulting perceptions. Still, advances in ‘pre-bunking’, which is understood as ‘inoculating’ people against disinformation so that they are better able to identify it, are important. While some studies have proven the benefits of pre-bunking, others say that its effects are much better in labs than in real-life settings (Buluc et al. 2023; Traberg, Roozenbeek and van der Linden 2022). There have also been important developments in the way voting takes place. SVAs, which include early voting, mobile ballot boxes, postal voting, proxy voting and telephone voting, have expanded the options people have to cast their ballots (International IDEA 2021). It will be important for researchers to study how the use of SVAs impacts both public and expert perceptions of electoral integrity.

Given this context, what drives people to mistrust their respective country’s election results and doubt the credibility of electoral processes, even without strong proof of malfeasance or error? What specific events, acts or issues do people consider to be most important when assessing the credibility of elections in their own country?

Up next