Political Finance in the Digital Age

Digital campaigning is on the rise across the world, and political parties and candidates spend more of their financial resources online. Other aspects of political finance—such as fundraising and reporting of parties’ and candidates’ finances—have also moved to the digital realm. While this trend creates new opportunities for reaching voters and for political participation, regulators are confronted with new challenges in their approach to this phenomenon. Existing regulatory frameworks are often insufficient to cope with the digitalization of campaign finance and need to be adapted. Yet many questions remain on how the rules should be changed and which regulatory approach should be followed.

This report offers an overview of the recent developments regarding digital campaign finance. Building on insights and findings from in-depth case studies—on Albania, Belgium, Bosnia and Herzegovina, Brazil, Chile, the European Union, India, Kosovo, Mexico, Montenegro, Nigeria and the United States—it discusses the main challenges faced. While these countries might differ in their specific political and electoral contexts, they experience very similar regulatory difficulties. Studying these cases enables the identification of best practices and possible lessons that can be drawn for political finance oversight bodies, civil society organizations, political party officials and legislators worldwide.

After an introductory section, the first part of this report discusses the current landscape of online campaign finance and identifies the main regulatory choices that exist: from a total ban on online campaigns, to incorporating the digital aspects into the rules on traditional campaigning, to the development of specific rules for online advertising. Regulators can opt for self-regulation or binding rules and can concentrate their regulatory efforts on political actors or online platforms. In the second part of the report, the most important issues in the regulatory framework governing online campaign finance are discussed. These include the full extent of the definition of digital campaigning, the differences in regulating online and physical campaigns, and the difficulties regarding official campaign periods and the concept of electoral silence.

The third and fourth parts of this report focus on the oversight of online campaign finance, with particular attention given to the role of monitoring agencies, civil society organizations, the media and online platforms. The fifth part goes into the main regulatory challenges, such as online third-party campaigning and in-kind contributions, cross-border campaigning and foreign interference, microcredits and cryptocurrencies, and the use of influencers and digital marketing firms.

The report concludes with a number of recommendations:

• The transparency of digital campaign finance should be increased, at the level of the online platforms as well as the oversight bodies, political parties and candidates. The further development of online advertisement repositories and transparency notices linked to online advertisements can improve the monitoring of digital campaign finance. Similarly, a wider definition of digital campaigning and a higher degree of granularity in financial reporting can improve oversight.
• Third-party campaigning should be included in the regulatory framework, by incorporating this into existing rules on campaign finance or by developing a specific new set of rules. Compulsory registration, the use of a dedicated bank account and financial reporting are only some of the provisions that can be considered.
• Better oversight of external media companies can strengthen overall campaign monitoring—for example, by setting up a public register of suppliers of external communication and public relations (PR) services for both online and offline political campaigns.
• Rules should also clarify the differences between ‘organic’ digital content and funded online political campaigns, and ideally create a level playing field between the digital and physical elements of electoral campaigns. The limits and possibilities of organic digital campaigning are an important point of attention in this respect.
• The capacity of oversight bodies should be strengthened to foster the specific expertise needed to monitor digital aspects of electoral campaigns. One important capacity-building strategy is the development of specific programmes to improve specialized knowledge and expertise in the control bodies, such as database management, data compilation and analysis, social media communication analysis, cybersecurity and information security, and financial transaction technology.
• It is also important to strengthen the capacity of civil society organizations, which fulfil a key role regarding oversight of digital campaign finance. The introduction of user-friendly and accessible tools for monitoring can also be helpful, as well as capacity-building programmes regarding tracking and monitoring online campaign finance, and the analysis of financial disclosure reports.

Politics is increasingly a digital affair: many aspects of political organization and electoral campaigning are shifting online. This is an inevitable part of the ongoing digitalization of society as a whole. In all countries in the world, Internet use is consistently on the rise. In 2024, global Internet penetration was at around 66 per cent, meaning that two-thirds of the world’s population have access to the Internet. In Western countries, Internet use stands at greater than 90 per cent (Kemp 2024). It is therefore not surprising that there is remarkable growth in the use of digital tools and social media platforms for political and electoral campaigning. Many have even argued that we have entered a new era—or ‘fourth phase’—of data-driven political campaigning, fuelled by the collection and analysis of large amounts of individual-level data with the purpose of microtargeting groups of voters (Gibson 2020; Kefford et al. 2023; Magin et al. 2017). Almost all 13 case studies analysed in this report show an increase in the financial resources that are spent on the digital aspects of campaigning, especially on social media platforms. The most striking example might be the USA, where it is estimated that more than USD 2 billion was spent on online advertising in the run-up to the 2020 elections (Venslauskas 2024). In Brazil, social media expenditure during election campaigns increased from approximately EUR 19 million in 2018 to EUR 69 million in 2022, largely at the expense of more traditional forms of campaigning (such as television and radio advertisements and sound trucks and cars) (Grassi 2024).

In some countries, more traditional forms of campaigning are still dominant, but most experts expect the digitalization of politics to spread everywhere. For example, in Belgium, investments in more traditional campaign tools are still in the majority, but the use of online advertising is on the rise. There are also substantial differences across parties in the country, with the far-right party and some liberal and socialist parties spending a higher proportion of their campaign budgets on online advertisements. Expenditure on digital campaigns from individual candidates, on the other hand, remains relatively limited (Vanden Eynde 2023).

India has been characterized by increased use of social media platforms for political advertising. The general elections of 2014 were the first time that big data analytics were used in political campaigns for the profiling and targeting of voters (Rao 2019; Sahoo 2024; Sen, Naumann and Murali 2019). The 2019 elections saw an estimated total campaign expenditure of USD 8.7 billion by all parties and candidates, and an estimated USD 8.3 million on online advertising on Google and Meta platforms. This relatively modest share, however, does not include the expenses of content development, or salaries or fees for social media staff or fees of the professional agencies. Since 2019, there has been a growing shift in the parties’ strategies towards using digital campaigning, which seems at least partly driven by the massive growth in Internet and social media users. More than half the country’s population now has access to a smartphone and social media channels (Sahoo 2024).

In some countries, the global pandemic of 2020–2022 has generated a profound transformation in political campaign strategies, since in-person and print media campaigns were substantially hampered (International IDEA 2020). Chile experienced five electoral campaigns in the period from 2020 to 2023, and social distancing measures were in force during much of this period, limiting the possibilities of physical campaigning. As a result, digital advertising—through social media platforms, automated calls, mass text messages and the use of artificial intelligence—replaced traditional forms of campaigning, such as rallies or billboards (Jaraquemada 2024; Olave 2021). In Montenegro, the 2020 parliamentary elections amid the Covid-19 crisis were marked by a tenfold increase in advertising costs on social media by political parties, compared with the previous elections, although these expenses are overall lower than for traditional media (Kovačević forthcoming 2025).

In most countries, Meta platforms seem to attract the bulk of campaign expenditure, although other online platforms are becoming increasingly popular. During the general elections in Chile in 2021, Meta platforms were used by all candidates for political advertising (Jaraquemada 2024). In Nigeria, Facebook/Meta is one of the most popular social media platforms and the preferred platform for accessing news online. It is therefore not surprising that most national politicians have a strong presence on the platform; as early as 2010, President Goodluck Jonathan of the Peoples Democratic Party announced his campaign to run for president on Facebook (Adetula 2024). In Brazil, Meta platforms (Facebook and Instagram) were the most popular for political campaigning, accounting for more than 80 per cent of social media expenditure by political actors in the 2018 general election and 2020 municipal elections. In the 2022 general election, this share dropped to approximately 60 per cent, with Google obtaining a higher share of political expenditure. Other social media platforms like TikTok and Kwai were not widely used (Grassi 2024).

Nevertheless, there are often specific features of a country that provide a unique context for its online campaigning. For example, in India, there has been a rapid expansion of social media applications in different regional languages, which are used by the political parties to specifically target certain subgroups in society (Sahoo 2024).

A further consideration is the extent to which digital campaigning differs in any fundamental way from more traditional campaign tools. On the one hand, both revolve around the same thing: communication from parties and candidates to voters in an attempt to secure their support. And in this respect, the pertinent question is how political actors differ in their use of resources to finance their campaign activities, and how this affects political competition more generally. Do digital technologies make campaigning cheaper and more efficient, benefiting smaller and less established parties and creating a level playing field, or can wealthier parties take advantage of their additional financial resources in the digital realm as well, thereby compounding pre-existing inequities (Gibson and McAllister 2015)?

On the other hand, there are important differences between digital and traditional campaigning. In some cases, a much larger audience can be reached through automation technologies. Compare online advertisements, automated calls and mass text messages with door-to-door canvassing and physical election posters. These new forms of campaigning often require fewer human resources and so create an advantage for parties and candidates that can rely on substantial financial resources but have fewer volunteers and/or members.

At the same time, in terms of reach, online advertisements are comparable with more traditional political commercials on radio and television. But the speed with which certain political messages are circulated online is often quicker, and this raises specific concerns with regard to the spread of political disinformation and fake news. In addition, it might become increasingly difficult for citizens to recognize something online as a political advertisement, since important information about the sponsor is usually missing. Similarly, the digitalization of political campaigning increases the possibilities for cross-border campaigns and thus also the risk of foreign interference, since it is less linked to the physical location of the person or organization running the advertisement.

The most important difference might be the wide range of possibilities for personalizing advertisements that social media platforms in particular offer. Targeted political advertising—especially microtargeting—relies on the collection and analysis of sensitive personal data to identify the interests and preferences of a specific audience, so as to be able to adjust the campaign message accordingly, with the aim of maximizing the influence on recipients’ voting behaviour (Bashyakarla et al. 2019). Such a level of personalization is not possible with more traditional forms of campaigning, like commercial billboards or television advertisements. In addition, political actors might engage in not only what Roemmele and Gibson (2020) describe as ‘scientific’ campaigns—aimed at informing and mobilizing voters—but also ‘subversive’ campaigns, with the objective of demobilizing voters and/or providing misinformation. The latter form of advertising creates fundamental democratic challenges. These problematic aspects of online campaigns might affect electoral integrity and have long-term implications after the elections (Agrawal, Hamada and Fernández Gibaja 2021).

In addition to the expenditure side of electoral campaigns, digital technologies are increasingly utilized by political parties and candidates for fundraising activities. On the plus side, these technologies offer new ways of campaign fundraising, such as microcredits or crowdfunding; however, they also bring new regulatory and oversight challenges. First, Internet platforms facilitate cross-border fundraising, which increases the risk of foreign funding and—consequently—the potential for foreign interference. Second, the use of digital technologies sometimes makes it more difficult to track financial transactions, especially when large-scale fundraising techniques, such as crowdfunding, are used. A particular problem in this respect is the emergence of cryptocurrencies—Bitcoin is probably the most well known. Such decentralized currencies bypass the traditional banking system, which creates important limitations for financial oversight (Venslauskas 2024).

Digital campaigning thus creates new challenges for regulators all over the world. One view of money in politics is that it is always looking for ways to break through the regulatory framework containing it, and the online environment can be considered one of the largest weaknesses in the overall system, with the potential to circumvent certain restrictions and limitations (Issacharoff and Karlan 1999; Power 2024). According to International IDEA’s Political Finance Database, less than 10 per cent of countries worldwide have specific provisions concerning online expenditure (International IDEA n.d.). Regulators need to be able to strike a balance between ensuring fair political competition and electoral integrity on the one hand, and respecting fundamental rights like freedom of expression on the other.

The objective of this report is to examine the main trends in online campaigning, how countries have, and have not, adapted to this new reality, which regulatory changes can be considered best practices and how the rules can be improved further to tackle the most important risks of digital campaign finance. This is done through the analysis of a number of instrumental case studies (on Albania, Belgium, Bosnia and Herzegovina, Brazil, Chile, the EU, India, Kosovo, Mexico, Montenegro, Nigeria and the USA), expert interviews and additional desk research.

The selection of case study countries was guided by several key factors. While regional balance was considered to allow for diverse perspectives and experiences, the selection prioritized countries where there have been notable progress or ongoing discussions on political finance reforms that address digital aspects, ensuring that the report captures emerging trends and regulatory efforts—the rationale why many of the countries are from Europe, a region which has made significant strides in this area. Lastly, the relative influence of each country within its region was considered, focusing on nations whose experiences and policies could serve as models or have a broader impact on regional discourse.

The case studies were developed between October 2021 and August 2024. While every effort has been made to include relevant regulatory updates across the 13 countries, the rapidly evolving nature of this field means that some recent developments may not be fully captured.

The rise of digital campaigning has created new questions and challenges for regulators, and authorities all over the world have taken different approaches, in particular on social media platforms (Figure 1.1). The first core question is whether or not a particular aspect of digital campaign finance needs to be—or can be—regulated. It is indeed perfectly possible to leave digital campaigning unregulated. This can be a deliberate choice based on certain normative considerations—such as is the case, to some extent, in the United Kingdom or USA—or because no consensus can be found among a majority of the political forces, sometimes because the use (and advantages) of digital campaigning differ substantially across the parties and/or candidates.

1.1. Banning digital campaigning

When regulators do opt to regulate digital campaigning, the most far-reaching option is to consider a ban. In some countries, social media campaigning has indeed been prohibited in its totality for political actors. This is often a country’s initial response to the emergence of this new communication instrument. For example, when the electoral management body in Chile drafted its first electoral guide on donations and expenses in the lead up to 2016 municipal elections, it included the provision that digital campaigning was forbidden (Jaraquemada 2024). In 2007, Belgian authorities added online advertisements to the list of prohibited campaign instruments during the official campaign period. The rationale behind the decision was that there were too many unknowns about the use and possible impact of this specific campaign tool (Vanden Eynde 2023). In Brazil, paid political promotions on social networks and search engines were banned in 2009 (Grassi 2024).

Yet, in all these cases, the initial ban was revoked at a later stage. The Chilean electoral management body had to quickly retract the ban on digital campaigning following criticism from the public and politicians (Jaraquemada 2024; Zamora 2020). In Belgium, the restriction on online advertisements was removed again in 2013, because regulators felt that this sharp distinction between offline and online advertisements could no longer be justified in the increasingly technological context of society (Vanden Eynde 2023). In Brazil, paid promotions for electoral campaigns on Internet sites and social media were also allowed again in 2017 (Grassi 2024). These practices show how a total ban, or near total ban, is difficult to maintain in practice. Denying political parties or candidates digital options in their campaigns is untenable, especially given the increasing global trend for political campaigns to move online.

1.2. Mainstreaming or innovation: Adapt existing laws or introduce new legislation?

A total ban on online campaigns is becoming increasingly scarce. Most countries are regulating—or considering regulating—the digital aspects of political campaigns. In this respect, regulators have held different perspectives on how digital campaign expenditure should be approached. Some took a ‘mainstreaming’ approach, in which online campaigning is considered a part of established political communication, albeit a new part. In terms of regulation, this means that the rules on online campaigning are incorporated into the mainstream: either the existing laws on political campaign finance have to be revised to include the digital aspects, or they are interpreted in new, innovative ways to enable them to encompass online campaigning as well. In either case, the existing, traditional legislation on campaign finance remains the regulatory core. A second ‘innovative’ approach considers online campaigning as a profoundly new and different form of political communication, which requires a fundamental reformulation of existing laws or else the creation of new rules that specifically target the digital aspects of political campaigns (Nieto-Vazquez 2023; Dommett 2020).

Examples of the mainstreaming approach include Chile, where the Constitution was amended in 2021 with a new provision on communication on social media platforms, designed to clarify the difference between freedom of expression and electoral propaganda. The new provision stated that, although political expressions made on personal or group profiles on social media did not constitute electoral propaganda and were protected by the right to freedom of expression, commissioning digital advertising services would be considered propaganda and would be subject to the rules on campaign finance (Jaraquemada 2024). Similarly, in Kosovo the mandate of the Independent Media Commission, as the oversight body for regulating traditional media during election campaigns, was extended to online campaigns. However, there is a gap in the regulatory framework with regard to digital campaigning, meaning that the commission’s ability to monitor and control online content is jeopardized (Cakolli forthcoming 2025).

In some cases, the existing regulatory framework is not well suited to dealing with online advertising. For example, in Montenegro, the media are defined in the Media Law as subjects engaged in creating and circulating media content that is aimed at an unspecified audience, and exercising editorial oversight or supervision of that content. This definition does not, therefore, cover social media—because of the lack of editorial supervision. Consequently, all rules that regulate the media landscape are not applicable to social media, and the media oversight body has no authority to monitor online platforms. In fact, there is currently no oversight body in Montenegro that monitors social media (Kovačević forthcoming 2025).

Latvia has taken a combined approach ensuring that the existing regulatory provisions for traditional campaigning apply also to online activities (mainstreaming), and creating new, specific rules for online campaign finance (innovation) (Cigane 2022).

The mainstreaming approach in Latvia shows that, if a legal framework addresses campaign finance in general in a comprehensive and robust way, it can also tackle the challenges linked to the digital aspects of campaigning more easily. This was nicely illustrated by Cigane (2022) in her discussion on how the existing campaign finance rules in Latvia could easily be mainstreamed to include online activities. For example, the general ban on corporate donations could also be extended to online publicity campaigns by companies on behalf of parties or candidates, since this constitutes an in-kind donation. Similarly, since parties must account for and report all campaign expenditure by their candidates, they are incentivized to keep the online campaigns of their candidates in check, as they are counted as part of the total campaign expenses from the party and risk exceeding the legal spending cap. A final example is the regulation of third-party campaigning: the rules in Latvia impose strict limitations on electoral campaigns by third parties, which makes undefined third-party activity online de facto impossible and linked to strict spending limits. (For a comprehensive overview, see Cigane 2022.) Consequently, the emergence of digital campaigning can also be considered a ‘stress test’ of the robustness of the existing regulatory framework on campaign finance.

With regard to the innovative aspect of their approach, Latvia also developed specific regulations related to online campaign activities in 2016. These rules stipulate that: (a) direct contact is required between a political party (or candidate) and a provider of advertising services (which prohibits any intermediation by a PR agency, for example); (b) all companies that offer advertising services, including online advertisement must provide a price overview to the oversight body before the electoral period; (c) companies that do not provide a price overview should be banned from publishing political advertisements during the electoral period; and (d) the sponsor of an online advertisement, must be clearly indicated (Cigane 2022). These measures were mainly aimed at increasing the transparency and accountability of online campaigning.

1.3. Self-regulation or binding rules?

In addition to the mainstreaming and innovative approaches, regulators also hold different perspectives on what the most suitable regulatory strategy would be. While some countries revise or develop binding rules on online campaign finance, others focus primarily on self-regulation—for example, by developing a code of conduct on digital campaigning—or even co-regulation, in which public and private bodies jointly create rules around a common goal (Heinmaa 2023).

In India, for example, there is no specific legislation on social media expenditure by parties and candidates, but the Election Commission of India—in collaboration with online platforms—developed a series of guidelines and a voluntary Code of Conduct to regulate online advertising. Among other things, they focus on the pre-certification of political advertisements, the disclosure of the candidates’ social media accounts when they file their nominations, and the disclosure of social media spending by candidates and parties. Although there have been some positive results arising from the presence of the code, there remain many difficulties in place with regard to effective oversight and enforcing compliance, particularly because of the non-binding nature of the guidelines (Sahoo 2024).

In the run-up to the elections of 2021, the Netherlands also introduced a specific Code of Conduct on online advertisements, to increase transparency, limit profiling and the use of intermediaries, and curb the spread of misinformation and hate speech. Of the 13 main parties and the major online platforms, 11 signed the code and generated relatively positive results (International IDEA 2021).

In Montenegro, parties were requested to sign a Code for Fair and Democratic Campaigns in the run-up to the 2023 parliamentary elections, building on similar initiatives in the past. The signatories committed to provide true and objective information, conduct factchecking and counter illegal influences and combat disinformation. Although the code showed potential, there were several breaches by political parties without severe implications (Kovačević forthcoming 2025).

In Bosnia and Herzegovina, the Press and Online Media Council developed a Press and Online Media Code to improve overall transparency and limit the spread of disinformation and hate speech, but it has generated mixed results (Micanovic forthcoming 2025). Overall, the use of self-regulation is increasingly considered insufficient to fully tackle all the challenges related to online political advertising. But it can be an important first step, especially in countries without a strong tradition of regulating political parties and campaign finance. An example of such a case is the Netherlands, where political advertising has largely been left unregulated and policymakers have traditionally counted on the parties themselves to uphold a culture of electoral integrity. This lack of an extensive regulatory framework made it more difficult to regulate online political advertising. Pending the introduction of new, binding rules, the Dutch Code of Conduct on the Transparency of Online Political Advertisements was negotiated between political parties and online platforms by International IDEA in 2021 as an interim measure (Heinmaa 2023; International IDEA and Dutch Ministry of the Interior and Kingdom Relations 2021).

1.4. Focal point of rules: Parties or platforms?

Similarly, countries can differ concerning which entities are the main focal points of their regulatory efforts. In line with existing frameworks, the rules on digital campaigning can either focus on regulating political actors, or opt to target the digital platforms that enable and host these new forms of political communication—just as traditional media may have been the object of regulation regarding political campaigning. The EU, for example, has developed far-reaching rules on the obligations of online platforms with regard to political advertisements, while its regulatory framework for political parties is more limited.

Yet, despite these different possibilities, many countries have not yet adapted the regulatory framework to incorporate the digitalization of campaign finance. One example is Nigeria, where the existing rules do not include any specific provisions on the role of digitalization and its impact on political and campaign finance. Social media remains to a large extent an unregulated space and has even been labelled a ‘free for all zone’ (Adetula 2024). While there is a specific provision in the Nigerian Constitution banning third parties from campaigning on behalf of a political party or candidate, this is not enforced in practice, including on social media: anyone can fund online advertisements with impunity. There are provisions against fake news and misinformation, but neither of these are enforced (Adetula 2024).

In some countries, the regulatory framework for campaign finance and online advertising can be rather complex, with different applicable legal acts, and responsibilities allocated across a wide variety of actors. For example, in Montenegro, regulation of the media sector alone (without campaign finance) comprises the Media Law, the Electronic Media Law, the Digital Broadcasting Law and the Law on Public Broadcasting Services of Montenegro, which adds to a certain level of regulatory complexity (Kovačević forthcoming 2025). These cases are a good illustration of how many countries still struggle to find a proper regulatory response to the ongoing digitalization of political campaigns.

The most important reason to regulate political campaigns is to create a level playing field among political parties and candidates, in order to ensure not only free but also fair elections (Agrawal, Hamada and Fernández Gibaja 2021). The regulation of digital campaigns with stable and effective rules is a difficult task for many reasons, including the high volume and pace of online communication, the tension with freedom of expression, and the need to enforce transparency and meaningful reporting from online platforms as well as political parties, candidates and third parties. A number of these regulatory aspects are the subject of intense discussion in many countries.

2.1. The definition of digital campaigning

The debate about the regulation of digital campaign finance immediately opens up the discussion on the breadth of the rules required: how should an online advertisement or digital campaign be defined, and which aspects of digital campaigning are—or should be—included in the regulatory framework? Online campaigning can take many forms and is constantly evolving, meaning that definitions are often quickly outdated and fail to capture new developments in the digital sphere.

The first important point to consider is the exact definition of an online political advertisement. Regulating political campaigns must be weighed against the importance of freedom of expression and not all online advertisements of a political nature fall within the scope of regulatory regimes. Civil society organizations (CSOs), such as Greenpeace or the World Wildlife Fund, might also launch advertisements online that are political in nature but these should not necessarily be covered by the regulatory framework for parties or candidates. At the same time, a definition that is too narrow risks creating loopholes that can be exploited by political actors to circumvent the rules.

Some regulators focus on a specific definition for online advertisements, while others include the digital aspects of advertising within a broader definition of all political or electoral communication. An example of the former is Mexico, where specially paid digital advertisements are defined as ‘insertions, e-banners, tweets, published messages, social media accounts, websites and other similar paid items whose purpose is to promote a campaign, a political party or a candidate’ (Nieto-Vazquez 2023: 4). This is quite a broad, encompassing definition, but does not include, for example, paid advertisements in favour of or against a specific policy issue (although this could be interpreted as a ‘campaign’). In the USA, the definition of a political advertisement is the subject of much discussion, since it includes advertisements ‘placed’ for a fee but not ‘promoted’ for a fee, which might mean that many advertisements fall outside of the scope of the rules (Venslauskas 2024: 9).

Other countries apply a definition that captures all electoral communication. In Chile, for example, electoral propaganda is defined as ‘any event or public demonstration and radio advertising, written, in images, in audiovisual media, through social networks, when there is a contract and a respective payment, or other analogous means, provided that it promotes one or more persons or political parties constituted or in formation, for electoral purposes’ (Chile 2016: article 30). This is again a rather broad definition, and means that any paid services on social media to promote a candidate or political party are considered political propaganda, and thus fall under the corresponding rules and transparency requirements. Similarly, a person, organization or company that is hired to develop or implement an online advertising strategy, or cases of telephone and email campaigning, also fall within the definition (Jaraquemada 2024).

However, even such broad definitions demonstrate definitional challenges: the Chilean law distinguishes between promotional activities and those messages ‘disseminating ideas or information on political acts’, which is considered a form of freedom of expression (Chile 2016). The instruction guides of the oversight body indeed also state that only ‘promotional advertisements’ that involve the commission of paid advertising services are considered electoral propaganda, and not the non-paid communication from the personal profiles of candidates and parties (SERVEL 2017, 2021). There nevertheless remains a legal grey area, and it is up to the oversight body to determine the difference and interpret the messages on a case-by-case basis. For example, when a social media influencer uses their personal account to promote a candidate or a party without specifically purchasing an advertisement, this can be considered either as a form of freedom of expression, or as ideas or information on political acts, or as an in-kind donation of services (comparable to the free performance of an artist during an electoral rally) (Jaraquemada 2024).

It is important to note that laws are not always limited to regulating paid political communication, and that the rules can be applicable only to political parties (or electoral coalitions) and candidates, or have a much wider scope. This is often related to the breadth of the definition of freedom of speech. In Mexico, for example, online political messages are divided into two categories: those messages without any other interest other than sharing views, and those published with the specific purpose of obtaining votes. In order to assess whether political communication falls within the scope of the rules, the regulatory authorities have adopted a set of criteria, such as the cost of the production involved, the timing of the message and the taxpayer status of the source of the message (Nieto-Vazquez 2023). Conversely, in Brazil, ‘organic’ electoral advertising—the publication of content without paid promotion—on social media is also regulated in the rules on electoral finance (Grassi 2024).

This touches upon the second important element to consider when defining digital campaigning: what should be included in the broader phenomenon of online campaign expenditure, and—consequently—become the object of certain obligations and restrictions. Rules on political campaigning differ substantially in terms of their range of regulated activities. India has a rather broad definition of social media expenditure that parties and candidates need to report. It includes payments for advertisements made to online platforms, operational expenses related to the development of campaign content or the maintenance of social media accounts, and salaries/wages paid to employees or professionals who undertake the tasks of maintaining these accounts (Sahoo 2024).

A non-exhaustive list of aspects of digital campaigning can include:

• the development and publication of websites and (micro)blogs (including domain specification);
• the creation of online content (copywriting, audiovisual material);
• political advertisements on websites and social media platforms;
• background research (surveys, big data analysis and so on) for profiling in preparation of (targeted) online campaigning;
• salaries (or fees) for data specialists or legal experts (to verify compliance with existing rules on the collection and use of data and other aspects of digital campaigning);
• salaries (or fees) for communication and social media staff, or influencers;
• fees for volunteers to disseminate and amplify political messages; and
• large-scale communication campaigns through email or instant messaging.

2.2. Different or similar rules for digital and physical campaigns

An important consideration is to what extent digital and physical campaigns have to be subject to the same—or at least similar—restrictions. This is linked to broader discussions about whether the digital aspects of political campaigns are best mainstreamed into the existing regulatory framework or a separate body of rules needs to be developed, and whether digital advertisements or campaigns should be included within a broader definition of political communication or not. Yet, in many countries, there are specific debates about the limitations that should be imposed on online campaigning, and whether these should be different from those on more traditional forms of campaigning. In Albania, clear limits on campaign expenditure in traditional media advertising exist, but these limits do not apply to social media advertising (Power 2024).

In Belgium, there has been an extensive debate on the regulatory limitations of social media campaigns. Since 1989, political campaigns have been closely regulated: there is a strict spending cap, and advertisements on both radio and television are banned, as well as the use of commercial billboards or commercial phone campaigns. In 2007, it was decided to ban online advertisements as well during the official campaign period (which at the time started three months before the election date), but not audiovisual content on the websites of parties and candidates (even when developed by an external media company). In practice, this meant that the digital toolbox for parties and candidates was limited to their own websites and emailing. In 2013, however, the ban on the use of online advertisements was removed (Vanden Eynde 2023).

However, there are ongoing discussions about curtailing digital campaigning again. In 2020, a new proposal was submitted to ban all commercial advertisements. The proponents argued that there is no valid justification to allow online advertisements, while political commercials in cinemas, on commercial billboards, and on television and radio are prohibited during the official electoral period. Another proposal aims to limit the share of campaign expenditure used for online targeted messages to 50 per cent of the total campaign budget of a party or candidate. The proposal is motivated by the need to protect personal data (since such targeted campaigns rely heavily on the collection of citizens’ personal data) and to limit the spread of fake news and disinformation (Vanden Eynde 2023).

Another example is the USA, where substantial differences existed concerning the transparency of online advertisements compared with traditional television and radio advertising. The latter has in recent decades been bound by strict and detailed rules, which stipulate that, among other things, clearly written, visible and understandable disclaimers are linked to each advertisement to disclose who paid for it and whether it was authorized by the candidate. Such disclaimers were historically absent from the regulatory framework for digital campaigning: it was not until 2022 that the Federal Election Commission introduced a similar rule on disclosure requirements for online advertisements (Venslauskas 2024). These examples show that a differentiated approach—in which online and physical or more traditional advertisements are regulated differently—potentially creates a tension for regulators, or can at least be the subject of intense debate.

2.3. Official campaign period and online advertisements

In several countries, a specific electoral campaign period is defined, during which time certain campaign activities are either permitted or, alternatively, subject to restrictions and limitations. Digital campaigning offers new challenges concerning the demarcation and enforcement of such a specific electoral campaign period, even if this is included in broader definitions of political communication and/or campaign expenditure.

A good example is Montenegro, which applies the concept of an ‘electoral silence’: a period of 24 hours before election day during which electoral propaganda via rallies and traditional media is prohibited. However, this provision is not applicable to campaign activities on social media (Kovačević forthcoming 2025). This is also the case in India, where the silence period is not applicable to online advertisements. During the 2022 elections in Brazil, there was a prohibited period for campaigning, but a difference was made between physical and online campaigning. Paid promotions—including on the Internet—were banned for two days before election day, and the promotion of printed materials was prohibited for one day before the election. Yet, what was not clear was to what extent organic campaigning—including non-paid advertising—on social media platforms was permitted during these silence periods (Grassi 2024).

Often parties and candidates try to work around the limitations of having a specific electoral period, and online campaigning can provide possibilities for doing this. In Nigeria, there are quite extensive rules on campaign expenditure (such as for advertisements, campaign materials and campaign rallies), but these rules are not in force outside the fixed electoral periods. There are clear indications that political advertisements on social media were widely used before the start of the official campaign period in the run-up to the 2023 presidential elections, as a way of circumventing the expenditure ceilings (Adetula 2024; Alayande 2022). Kosovo is also characterized by similar evidence of frontloading online campaign expenses so as to stay below the expenditure threshold linked to the official electoral period (Cakolli forthcoming 2025). In Bosnia and Herzegovina, the ban on paid advertising before the official electoral period is frequently circumvented by parties and candidates, especially through social media campaigning (Micanovic forthcoming 2025).

Another question is about the extent to which the costs of developing advertising content should be included in the total campaign cost if the advertisements were created before but published during the regulated electoral period. This is, of course, particularly important where electoral spending caps are in place.

Belgium also has an official campaign period—four months before the elections—with a clear expenditure ceiling: political parties may not spend more than EUR 1 million on their electoral campaigns (and different expenditure caps apply to individual candidates depending on their position on the electoral list). However, the main effect of this regulation has been an enormous increase in expenses on political advertisements before the official campaign period (Maddens et al. 2024). By way of a response, new proposals are being discussed in Belgium to introduce an annual spending cap of EUR 1 million per political party on all political dissemination and advertisements (both offline and online). According to the proponents, this would limit the enormous amounts of public money that go to social media platforms—Belgian parties are predominantly financed through state subsidies—and would create a more level playing field between parties with and without substantial financial resources (Vanden Eynde 2023).

Monitoring campaign finance has always been a challenging task, and the expansion of digital campaign tools has only complicated it. In terms of the definition of oversight, it is useful to differentiate between primary and secondary oversight. Primary oversight entails the monitoring activities of the bodies that are assigned a specific formal role in the regulatory framework, such as audit companies or election monitoring bodies. Secondary oversight is related to the activities of CSOs, activists, journalists or academics, which follow and scrutinize the campaign activities of political actors.

3.1. Role of oversight agencies

Oversight agencies are heavily dependent on the responsibilities and boundaries that are set by the general regulatory framework on campaign finance. In most European countries, no authorities have a clear formal oversight mandate for online political advertising (Heinmaa 2023). Regulators have different options on how to deal with the digitalization of campaign finance. Either they take a more conservative position and act as ‘constructionists’ that work from what is written and (only) do what is explicitly stipulated in the rules, or they take a more ‘activist’ position and treat the existing electoral and campaign finance rules as a living document that can be interpreted and adjusted to changing political circumstances (Power 2024). Even if a clear regulatory framework is not present, this does not always prevent an oversight body from acting. For example, in Brazil the main oversight entity—the Superior Electoral Court—regulates elections and electoral expenses mostly through resolutions, in the absence of any specific legislative acts (Grassi 2024). When regulatory frameworks are adapted to incorporate digital campaigning, new powers are often entrusted to existing oversight entities. For example, in Mexico the Constitution was amended in 2007, which ushered in a new era of political communication, giving the electoral authorities the power to investigate and sanction organizations and individuals both inside and outside politics (Nieto-Vazquez 2023).

One important question is whether oversight agencies should be given the power to proactively monitor political messages before they are published. For example, in Mexico, the constitutional reform of 2007 granted the National Electoral Institute the authority to review and approve campaign messages before they are broadcast, on both traditional and online media: parties and candidates must submit a proposal for any paid audiovisual advertisement (Nieto-Vazquez 2023).

The proper monitoring and oversight of online campaign finance requires specialist knowledge and expertise, such as database management, data compilation and analysis, social media communication analysis, cybersecurity and data security, and financial transaction technology. This often means that the oversight agencies need to develop or hire new expertise to complement their existing capabilities. The Irish Election Commission can—in addition to its own investigation of online advertisements—also engage external assistance and expertise (Heinmaa 2023). Similarly, in several countries, oversight agencies employ observers in the run-up to an election to help in the monitoring of compliance with campaign finance rules. However, while observers are well established for physical campaign activities in some countries, there are fewer observers in place for digital aspects of campaigning. Properly monitoring online campaigns also requires specific skills, which can mean additional training is needed. For example, in Nigeria the Independent National Electoral Commission deployed observers in the run-up to the 2023 general elections. While they were relatively effective at tracking expenditure on large billboards, posters and advertisements in traditional media, they were less successful in monitoring the online advertisements of parties and candidates (Adetula 2024).

3.2. Prerequisites for successful oversight

A first important condition for successful oversight is the availability and quality of information on the actual digital expenditure incurred by political parties and candidates, in particular the level of detail (or granularity) of that information. In this respect, the difference between the information available for primary scrutiny and that for secondary scrutiny is pertinent. While CSOs or journalists depend on data that is made publicly available (from the parties and candidates themselves as well as the official bodies or even online platforms), official oversight entities generally have more possibilities for accessing information. Yet, the level of detail and the timing of the information on expenditure often depends on the provisions included in the regulatory framework.

A couple of examples can illustrate this point. During the 2022 elections in Bosnia and Herzegovina, the oversight body obliged parties and candidates for the first time to include in their campaign expenditure reports a separate line covering online spending. This increased transparency, although to a limited degree, but the lack of detail means it is impossible to determine exactly how, and on what platforms, the money was spent (Micanovic forthcoming 2025). In the USA, there are no regulations that require sponsors of online advertisements to disclose their expenses and their donors (Venslauskas 2024), which makes comprehensive secondary oversight difficult, if not impossible. In many countries, the categorization used in the financial reports is too vague to allow for a detailed assessment of online communication expenditure (even for the official oversight bodies). In Nigeria, there is only limited itemized detail in the financial reports of the parties and candidates, which hampers proper oversight (Adetula 2024). Similarly, the absence of a dedicated category for online campaign expenditure on the financial reporting forms in Kosovo substantially hampers proper oversight. It can lead parties to categorize these expenses under different headings and therefore makes it difficult to compare and, more importantly, to enforce the legal threshold of EUR 5,000. ‘In addition, since all expenditure and donations need to be done from a dedicated party bank account, candidates are consequently not allowed to receive donations directly, or make personal campaign expenses. However, the lack of financial reporting from these actors makes proper oversight over these actors more difficult’ (Cakolli forthcoming 2025). In Montenegro, parties and candidates are required to provide comprehensive reports on their campaign expenses to the oversight body, but it is very difficult to know the exact cost of digital campaigns. The reports indicate the overall campaign costs, but many parties enlist media companies to conduct online campaigns on their behalf, and the exact expenses of these advertisements are not indicated (Kovačević forthcoming 2025).

The Indian oversight body also adapted the template for the financial accounts in 2022 to include a new column specifically dedicated to expenditure on digital campaigning (Sahoo 2024). Similarly, in Chile the electoral management body—the Electoral Service—has developed a specific form for the reporting of all expenses specifically related to digital campaigns. This includes, among other things, information on the platform that is used, the contract period and the amounts involved. However, not all political parties and candidates used the form to report their digital expenses, and their financial reports lacked details on the specifics of their digital campaign expenditure (Jaraquemada 2024). In an attempt to tackle this issue, new rules now oblige online platforms to publish the fees they charge for all advertisements (including political advertisements), and to report on the advertisements that are sponsored by political parties and candidates. These new provisions are also enforced, as the 2023 sanction of Google by the Electoral Service for non-compliance can testify.¹ Another example is Latvia, where the mainstreaming approach has resulted in all aspects of online campaigning being included in the itemized and detailed political finance reports (Cigane 2022).

As important as the level of detail, in facilitating the task of oversight, is the ability to check the information provided by the parties and candidates against the original documents and other data. In Mexico, parties and candidates must keep all contracts and invoices for advertisements published on the Internet (Nieto-Vazquez 2023). In India, the oversight body expects all candidates and political parties to keep information on all aspects of online campaign spending, such as direct expenditure on the publication of social media advertisements, payments to Internet companies and websites, operational expenditure on creative content development, and salaries paid to professionals/staff (Sahoo 2024). In this respect, the use of an electoral reporting system facilitates the monitoring process (Wolfs 2024). In Brazil, parties and candidates must use the electronic reporting system to notify the oversight body of the amounts spent on digital campaigning, including a description of the expenses, information on the service provider and a copy of the invoice or receipt (Grassi 2024).

Especially when monitoring online campaigns, the ability to compare the information provided by the parties and candidates with data originating from the social media platforms themselves is incredibly useful. In Montenegro, there were many cases of contestants initiating social media campaigns before they had established a dedicated campaign bank account, which complicated proper oversight of campaign expenditure. In addition, there is only limited access to data from the online platforms, which makes it difficult to compare this with the financial reports of the parties and candidates to assess their accuracy (Kovačević forthcoming 2025). Yet when data is provided by the platforms, but it is not comprehensive, this creates new oversight challenges. For example, the digital spending data of the Chilean parties and candidates did not match the data published by Meta—the most widely used platform, Facebook—which might indicate that parties and candidates also reported expenditure on other social media platforms (for which no data sources were available) (Jaraquemada 2024), or could mean they made errors in their reports, whether accidental or intentional.

A second important condition for successful oversight is that the monitoring bodies are given the instruments to duly fulfil their task. Indeed, a comprehensive regulatory framework is not sufficient to ensure fair political competition. This also requires a strong oversight entity that has the competences, resources and willingness to enforce compliance with the rules.

There are many examples of politicians that have breached the rules but did not encounter any repercussions. In the run-up to the 2022 elections in Brazil, a non-governmental organization concluded that 171 of the 375 social media accounts (almost half at 46 per cent) did not respect the prohibited period for paid electoral campaigning, and 306 of the 375 accounts (82 per cent) did not comply with the ban on organic campaigning in the last days before the elections (Grassi 2024). Similarly, 20 per cent of the first-round candidates did not declare their expenses, mainly because the current rules relieve non-elected candidates from reporting obligations (Grassi 2024).

This has also been the case in Chile, where rules on the permitted periods for electoral campaigning were not respected on social media in the context of the 2021 elections. The oversight body experienced substantial challenges to enforcing compliance (Jaraquemada 2024). Similarly, while the Nigerian Constitution bans anyone other than political parties and candidates from political campaigning (on behalf of parties and/or candidates), this is not enforced, especially on social media. In practice, anyone can place paid advertisement on social media in support of or opposing a political party or candidate. When funded by third parties, this expenditure often remains undisclosed (Adetula 2024). In addition, Nigerian parties and candidates often neglect to provide detailed information on campaign expenditure (although this is legally required), but there has never been a sanction for political finance infringements in the country. For example, after the 2023 general elections, no financial report on the campaign expenses of political parties had been submitted to the oversight body within the required timeframe, and when they were later submitted, important information (especially on the expenditure of individual candidates) was missing (Adetula 2024).

3.3. Role of civil society and the media

Civil society has always had an important function in monitoring elections, as CSOs conduct the core of the secondary oversight that complements the work of the official bodies. The shift of campaign activities to the digital realm has only increased the importance of their role. During the 2022 elections in Brazil, the Getulio Vargas Foundation think tank monitored the activities of political social media accounts and found several breaches of the campaign finance rules (Grassi 2024).

In countries where there is already limited involvement in monitoring campaign finance by CSOs, extending their work to the digital aspects of political campaigns can be even more challenging. In Nigeria, the engagement of CSOs in monitoring party and campaign finance is rather marginal and uncoordinated, mainly because of limited resources and organizational capacity, and because of limited collaboration from the political actors themselves (who refuse to share financial information) (Adetula 2024). Albania has a vibrant civil society that is active in the field of monitoring campaign expenditure, but most of these organizations are donor-funded, which makes them fragile and vulnerable to declining funding from international donors, making their work unsustainable and subject to funding availability (Power 2024).

Even in countries with strong oversight bodies, CSOs fulfil an important function. The Corruption Prevention and Combating Bureau in Latvia, for example, is one of the strongest oversight entities in Europe (and the world), but still relies on citizens to supply the majority of the evidence it evaluates on possible campaign finance violations, including in the online space. This high degree of involvement from civil society can be explained by the serious level of attention paid to campaign finance in the media and society in general, and by the user-friendly reporting procedures. The latter were developed as a direct response to possible foreign election interference online in the run-up to the 2018 elections, following a call for action from civil society (Cigane 2022). This decentralized approach, counting on the collaboration of citizens and CSOs, can be particularly useful in an online environment that is characterized by the microtargeting of small groups in digital political advertisements that might escape the attention of the official entities. For this reason, the Election Commission of India made specific investments to strengthen the involvement of CSOs and even ordinary citizens in its monitoring process. It developed an application—the ‘cVIGIL app’—which allows citizens to report potential instances of misuse of campaign finance. Despite the potential of such an application, the limited user-friendliness and minimal response from the oversight body limit its effectiveness (Sahoo 2024).

Online platforms play a central role in digital campaigning, and their actions can hamper or strengthen oversight, especially in countries where dedicated legislation is lacking. The various online platforms have had different approaches towards political advertising. Several—such as LinkedIn, Telegram and TikTok—prohibit all forms of political advertising: sharing political beliefs and expressions is only allowed as organic content. Following several controversies regarding political campaigning, X—previously Twitter—banned political advertisements, but announced it would lift the ban again in the run-up to the 2024 US elections (Venslauskas 2024).

Other platforms have taken a more pragmatic approach, especially because political advertisements constitute an important source of revenue. Obtaining cooperation from social media platforms can be challenging, especially in smaller countries, and sometimes requires substantial political pressure. For example, Latvia undertook several public diplomacy activities—including the country’s president making a visit to the headquarters of Facebook in California—in an attempt to gain the collaboration of the platform in the run-up to its elections in 2018. Yet, this did mean that the country was able to obtain all the requested information regarding campaign spending on the Facebook platform, and a comparison between the data from the platform and the financial reports from the political parties showed a high level of similarity. In addition, from 2019 onwards, the platform only allowed political advertisements from users residing in the country where the elections were held (in an attempt to curb possible foreign interference) (Cigane 2022). An additional regulatory challenge is that these platforms might offer services in a country without being legally established there, which may result in legal challenges and disputes regarding whether a country’s electoral body has any jurisdiction or control over them.

The first important consideration is the level of transparency that online platforms can provide on the sponsors of political advertisements. Several platforms have started proactively providing more transparency on political advertising, often through a voluntary code of ethics. In the run-up to the 2019 European elections, the main online platforms developed a Code of Practice in collaboration with the European Commission to provide more transparency on political advertising and to tackle disinformation. However, the results were, for the most part, disappointing (Wolfs 2024).

Also in 2019, the major social media platforms that were active in India—Twitter, Facebook, Google, ShareChat, TikTok and WhatsApp—drafted a similar code in collaboration with the Internet and Mobile Association of India. The Voluntary Code of Ethics contained four key commitments: (a) implement education and awareness campaigns, and establish a fast-track grievance redress channel to take action on objectionable posts; (b) take action within three hours of reported violations of the mandatory 48-hour period of no-campaigning and advertising silence before voting; (c) the pre-certification of all political advertisements published on their platforms by the government’s media certification and monitoring committees; and (d) report paid political advertisements and label them accordingly (Mehta 2019; Sahoo 2024). However, the code only seems to have had a limited effect, as there have been widespread instances of violations of the code and only limited follow-up by the oversight bodies. For example, in the context of the 2019 elections, 510 code violations were reported but only 75 were analysed by the Election Commission and no penalties were imposed (Sahoo 2024).

An important challenge in this respect is that the platforms do not always apply the same definition of political advertising, or they use one that deviates from the formal legal definition. This has important implications for the number and type of advertisements that are included in the repositories of the platforms, and thus also the possibilities for oversight entities and other actors to monitor social media expenditure.

The definition of political advertising used by Meta is anything that: (a) is made by, on behalf of or about a candidate for public office, a political figure, a political party or a political action committee or advocates for the outcome of an election to public office; or (b) is about any election, referendum or ballot initiative, including ‘get out the vote’ or election information campaigns; or (c) is about any social issue in any place where the advertisement is being run (Jaraquemada 2024). This last condition in particular substantially widens the scope of the definition and is broader than the legal definitions used in most countries. The broad definition Meta uses also makes it a lot more difficult to differentiate between ‘political’ and ‘electoral’ expenditure. This is complicated further by the fact that it is difficult to delineate specific time periods in the Meta advertising archive for collecting historical expenditure data. In Canada, the electoral rules differentiate between two types of advertisements that online platforms must include in their online registries: ‘partisan ads’ focused on promoting or opposing a party or candidate; and ‘electoral ads’ that take a specific position on any issues during a federal electoral campaign (Power 2024).

Google has a different policy for political advertising depending on the region. In some regions, advertisements may only be published if the advertiser is verified by the platform. In regions or countries where electoral advertising requires verification, the advertisement must contain information on the sponsor: an advertiser identification label ‘is generated from the data provided during the verification process and is automatically included in most ad formats’ (Jaraquemada 2024 ).

Without detailed information from the platforms themselves, it is often difficult—and in many cases even impossible—to track the donors of online advertisements, which further complicates verification procedures. This can then have an impact on assessing whether parties and candidates have complied with any spending limits that might apply. Data from online advertisement repositories can indeed provide an important source of comparison with the information provided by the parties and candidates themselves in their financial reports and may lead to further investigation in the case of substantial inconsistencies. This is why several countries have taken binding regulatory action to impose transparency requirements. For example, in Brazil all online electoral advertisements must be identifiable by the candidate’s corporate taxpayer number as well as information on the party or electoral coalition. In the USA, transparency disclaimers on the sponsor were absent from online advertisements for a long time, while being mandatory for advertisements on traditional media. The Federal Election Commission introduced new rules, taking effect in 2023, requiring each online communication to have a disclaimer which ‘must be presented in a clear and conspicuous manner to give the reader, observer, or listener adequate notice of the identity of the person that paid for the communication’ (Federal Election Commission n.d.). More transparency measures have been proposed—such as a ban on advertisements funded by foreign nationals and the mandatory creation of databases of online advertisements—but these have not (yet) been approved by the US Congress (Venslauskas 2024).

Yet, even when online platforms provide transparency on advertising costs, the information is often incomplete or inaccurate. The repository for Meta includes banded estimates of expenditure only, and because of the nature of online advertising, including its virality, the use of influencers and the relation between organic and paid content, the costs of different advertisements can vary substantially, with prices rising and falling (Dommett and Power 2019; Nadler, Crain and Donovan 2018; Power 2024).

The EU has also introduced far-reaching legislation on transparency. It stipulates that advertisements on online platforms should be clearly labelled as such to allow users to differentiate them from other content. In addition, an advertisement must identify both the natural or legal person that paid for it and the natural or legal person on whose behalf it is presented. Applied to political advertising, this means that both the sponsor of a political advertisement and the party or candidate that it is promoting must be made clear. Consequently, these provisions also entail de facto transparency in online campaigning by third parties (Wolfs 2024). From 2025 onwards, every political advertisement should be linked to a ‘transparency notice’, which has to include: the identity and contact details of the sponsor; the period of publication; the amounts spent and the source of funds; and the election or referendum to which the advertisement is related. In addition, for online advertisements, it must be made clear whether an advertisement is specifically targeted, and if so what criteria have been used for that targeting, and whether and to what extent amplification techniques were used to boost the advertisement’s reach. Platforms will not be allowed to publish the advertisement if it does not include this information (Wolfs 2024).

The EU approach seems to be heavily inspired by Ireland’s Electoral Reform Act of 2022, which also obliges the use of a digital imprint for online political advertising. The text ‘political advert’ should be clearly displayed, and a transparency notice should be accessible, which should include details on the buyer of the advertisements, and on the use of microtargeting, the total amount paid for the advertisement, the estimated size of the audience, the number of views and a link to the online advertisement repository of the platform. In addition, the act also obliges platforms to verify the identity of the sponsors of online political advertisements (Power 2024).

Discussion on the role of online platforms often goes further than providing mere transparency; some regulators also expect a level of intervention by the platforms, especially in the context of reducing possible fake news. However, the debate on countering fake news and disinformation touches upon the important principle of freedom of speech. Removing content—in this case, political content—from a social media platform must always be weighed against the freedom of expression of citizens. That is the reason why in many countries only a judicial authority can impose the removal of online content. But the slow pace of a judicial process often contrasts with the volatility and speed of circulation of possible fake news on online platforms. Regulators are trying to adapt to this new environment. For example, in Brazil, the rules were changed in 2022 to grant judicial authorities more possibilities in tackling disinformation. Previously, a separate lawsuit had to be filed for each link that was suspected of containing fake news, even if different links had the same exact content. The new rules allowed judicial authorities to target any existing or future online links that contained the same content (Grassi 2024).

The digitalization of political campaigning has created new challenges for regulators, and further complicated existing challenges.

5.1. Third-party campaigning and in-kind contributions

Third-party campaigning has long been a sensitive issue for regulators, but the new forms of digital communication have only amplified the challenge. The term ‘third party’ refers to stakeholders, other than the political parties and candidates themselves, who engage in electoral campaigning. Some of the most well-known third parties are the political action committees (PACs and Super PACs) in the USA—organizations that pool donations and contributions with the aim of funding campaigns for or against certain candidates or policy issues.

Third parties can be influential actors in an electoral competition. The USA is a typical example here, but research on the 2019 elections in India also indicates that most expenditure on digital platforms was not made by the political parties and candidates but by their sympathizers or affiliated groups. Since political advertisements by parties and candidates need to be pre-certified, third-party campaigning can be used to circumvent this requirement. In addition, the parties and candidates often also escape liability for the content of these third-party posts because of weak regulatory provisions (Mehta 2019; Sahoo 2024). In Brazil, companies are no longer allowed to contribute to political campaigns, so instead they hire people to campaign on behalf of the parties or candidates or pay for online campaigns on Meta platforms or TikTok. For example, there was evidence that entrepreneurs had spent millions in mass messaging services in favour of Bolsonaro during the 2018 presidential campaign (Campos Mello 2018). This constituted an illegal strategy, since Brazilian law prohibits companies from donating to political campaigns (Grassi 2024), but the direct nature of such campaigning could have been designed to circumvent the restrictions on third-party campaigning.

Third-party involvement also appears to be on the rise in Kosovo, where for example one-third of campaign advertisements on Facebook and Instagram during the 2021 local elections were attributed to third parties. Interestingly, these accounts were mainly focused on negative campaigning and attacking certain candidates instead of placing advertisements in support of a specific party or candidate(s) (Cakolli forthcoming 2025). In India, the Bharatiya Janata Party (BJP) deployed more than 1.2 million ‘volunteers’ (actually paid by the party) on social media, which in some cases operated as ‘troll armies’ to dominate the narrative online and push the party’s political platform (Sahoo 2024; Sen, Naumann and Murali 2019). In general, third parties are often found to be spending much more than official political parties and candidates on online advertising globally. Their substantial role in elections thus highlights the importance of regulating third-party activities.

In some countries, an important difference is made between paid and unpaid advertisements from third parties. If they launch paid advertisements, this is often included within the scope of rules on campaign expenditure. In Belgium and Mexico, for example, any campaign expenses from third parties are added to the total expenditure of the political parties or candidates being promoted. Of course, such a measure only has an impact when expenditure limits are in place. In 2018, the Canada Elections Act was revised with the specific aim of establishing a spending ceiling for third parties, banning the use of foreign funding by third parties and increasing the transparency of their participation in the elections (Agrawal, Hamada and Fernández Gibaja 2021).

The Chilean definition of electoral propaganda is sufficiently broad to also include third-party campaigning. It is defined as ‘any public event or demonstration, or radio, written, visual, audiovisual or other similar media advertising, if it promotes one or more persons or political parties, for electoral purposes’ (Chile 2016), which means that regardless of who pays for online political advertising, it falls within the rules when a candidate or party is promoted in the context of elections. Moreover, even if a person offers their services for free—for example, for the development or implementation of an online advertising strategy—these must be declared and valued as campaign contributions in line with the respective rules (Jaraquemada 2024). However, advertising and propaganda is increasingly being seen from campaigning movements, grassroots or social organizations that campaign specifically in favour of, or against, issues that are closely related to the political platforms of parties or candidates. While this could fall within the general right of freedom of expression, research by investigative journalists has shown how, in some cases, these organizations were used to issue electoral propaganda. Yet such campaigns have largely escaped monitoring by the oversight body (Jaraquemada 2024).

In Nigeria, third-party campaigning is banned altogether, since the Constitution limits political campaigning to the parties and candidates themselves. However, especially online, many individuals and organizations launch political advertisements in support of a party or candidate, but this remains unchallenged by the regulatory authorities. In addition, the costs for these online advertisements by third parties remain predominantly undisclosed (Adetula 2024), posing additional challenges for the transparency of campaign finance.

5.2. Instant messaging applications

An often overlooked element of electoral campaigning is the use of instant messaging applications, such as WhatsApp and Telegram, in electoral campaigns. Although most of these applications do not allow paid advertising, they are becoming an increasingly important campaign tool in many countries. For example, during the 2018 presidential elections in Brazil, there was a great spread of messages supporting candidate Jair Bolsonaro on WhatsApp. This led researchers to believe that so-called ‘super participants’ were involved—that is, a limited number of individuals responsible for a very high number of posts—which gives the impression of a coordinated campaign financed by the Bolsonaro campaign, albeit indirectly by sympathetic individuals or groups (Grassi 2024). There was also evidence that entrepreneurs had spent millions in mass messaging services (Campos Mello 2018). This was a form of third-party campaigning and illegal, since Brazilian law prohibits companies from donating to political campaigns (Grassi 2024).

Similarly, an analysis of the 2018 elections in Malaysia hypothesized that the campaigns through WhatsApp played a more important role than those on Facebook in terms of influencing voting behaviour (Nizah and Bakar 2019). Research on electoral campaigns in Spain has shown how political parties have used Telegram intensively, partly in response to WhatsApp blocking political parties from being able to send mass messages in the run-up to elections. The platform was mainly used as a way of disseminating the party manifesto, as a mobilizing tool to ask citizens to vote for them and as a negative campaign instrument criticizing political competitors (Alonso-Muñoz, Tirado García and Casero-Ripollés 2022; Tirado Garcia 2023). The particular risk of spreading disinformation through such instant messaging applications has been strongly acknowledged by scholars (Rossini 2023). It is clear that the use of these instant messaging applications creates additional regulatory challenges: while content on other social media platforms, such as Google, YouTube or Meta platforms, is in the public domain and can be monitored by open sources, this is not the case for services like WhatsApp which uses encrypted messages.

5.3. Cross-border campaigning and foreign interference

Digital campaign tools make it easier to conduct cross-border electoral campaigning, which can jeopardize the level playing field for political parties and candidates, or even increase the risk of possible foreign interference. There are indeed strong indications that foreign actors have at least attempted to manipulate democratic elections in the EU by deepening polarization and eroding public trust in the elections (Albrycht et al. 2021; Dunčikaitė, Žemgulytė and Valladares 2021). This is also the case for other parts of the world. For example, during the 2021 presidential elections in Chile, one of the first-round candidates was not physically present in the country during the pre-electoral period (due to unresolved legal issues), and consequently had to coordinate his campaign from abroad, mainly using digital campaign tools. However, since national finance regulations and oversight rules often only apply within a country’s territory, this creates regulatory challenges for the oversight body to control campaign expenses incurred in another country (Jaraquemada 2024).

In Montenegro, there have been cases where election campaigns were moved to the neighbouring country, Serbia, in an attempt to bypass Montenegrin electoral laws (Kovačević forthcoming 2025). Especially for smaller countries, such as Montenegro, that share a language with neighbouring countries, the borders of the country are not the borders of the information space: several media companies operate under a Montenegrin licence outside of Montenegro. The use of online platforms makes such a strategy even more feasible: the biggest online news media are stationed outside the country. Indeed, there have been several instances where undisclosed foreign funding was used on Facebook on campaigns from unregistered pages (Kovačević forthcoming 2025).

Some platforms have introduced measures to prevent—or at least limit—the possibility of cross-border campaigns. For example, in the run-up to the 2019 elections for the European Parliament, Facebook only allowed paid political advertisements for organizations that had a registered address within the particular country. This made cross-border campaigning on the platform practically impossible (Wolfs 2022). While this could be an important measure to prevent foreign interference, it is difficult to develop a regulatory—and particularly monitoring—environment that can entirely prevent the attempts of foreign entities to influence elections through social media campaigns. Some platforms have only introduced minimal rules, which also amplifies the risk: there are, for example, clear indications that Russia has used TikTok and Telegram to influence the 2024 elections in Moldova (Gavin 2024).

Ireland has introduced strict limits on online advertisements purchased from abroad: only Irish citizens or a company that has a registered office in Ireland can buy such advertisements. The only exception is for EU citizens from outside of Ireland who want to buy advertisements related to the European Parliament elections (Power 2024).

The risk of foreign interference was also an important driver for improving the oversight of online campaigning in Latvia in 2018. The country held parliamentary elections that year, and CSOs and journalists raised the alarm about potential foreign influence in the elections and cast doubt on the overall preparedness of the monitoring agencies. Specifically, there were concerns over anticipated microtargeting on social media, as a way of distributing disinformation and undisclosed political advertisements, as well as possible attempts to circumvent campaign restrictions through divisive online issue-based campaigns. As a response, a special taskforce on election security coordination was created, more intensive collaboration with the main social media platforms was developed, and the oversight body invested in new campaign spending monitoring tools and more user-friendly notification procedures for potential violations of the rules (Cigane 2022).

5.4. Influencers and digital marketing firms

More and more political parties and candidates rely on influencers, external experts and digital marketing firms to develop or support their electoral campaigns. Social media in general and influencers in particular are an increasingly important source of information and news for citizens, in particular younger cohorts. For example, a survey conducted in 2022 in Argentina, Brazil, Colombia and Mexico found that this is the most important channel of information on politics for young people aged 16 to 24 (Luminate 2022).

The recruitment of influencers can take several forms. First, they can be hired to endorse or campaign on behalf of specific candidates or political parties. In this respect, recruiting influencers (such as performing artists, television presenters or radio commentators) is a common strategy in Brazil (Grassi 2024). It is also a regular practice in the USA. Small-scale influencers or ‘nano-influencers’ (with fewer than 10,000 followers) are often viewed as ordinary individuals, which conveys a certain authenticity. However, most of the financial transactions between the political parties, candidates or third parties on the one hand and the influencers on the other hand remain hidden: online platforms are unable to control any deals that are made off platform and concealed agreements are often used (Venslauskas 2024).

Second, influencers can be used to conduct negative campaigns focused on portraying political opponents in a bad light. In Nigeria, paid social media influencers play an important role in political campaigns. This has received substantial criticism, as they have been considered responsible for the circulation of a large volume of disinformation and other strategies—such as hashtag manipulation, deliberate mistranslation, impersonation, and manipulating audio and video material—with the sole aim of steering the elections in a certain direction (Adetula 2024; Okoro and Adibe Nwafor 2013).

Finally, influencers are also increasingly taking the step of becoming politically active themselves and competing for public office, in an attempt to translate their (sometimes) large support base of followers and supporters into political power. In Brazil, the candidacy of influencers sometimes takes the form of ‘vote pullers’—they use their popularity not only to get themselves elected, but also to help other candidates gain a seat by attracting a maximum number of votes for the entire party list.

In many countries, employment as a social media influencer for a political campaign can be appealing. For example, in Nigeria, the best-known political influencers hired by political parties and candidates can earn up to NGN 500,000 (or USD 1,090) a month. Some influencers are also rewarded with lavish gifts, government contracts or political appointments. The spread of false narratives and disinformation is often their primary task. In addition, this expenditure on influencers often remains unreported by the political parties and candidates (Adetula 2024). In the USA, disclosed data shows that a Democratic PAC paid between USD 300 and USD 500 per post to TikTok influencers, while other influencers were simply hired as part of the fixed campaign staff (Venslauskas 2024).

Yet, even when influencers are not paid by a party or candidate, rules on campaign finance might still apply. In Chile, any form of advertising to promote a candidate or party for electoral purposes is covered by the campaign finance rules. While the dissemination of ideas on political acts by natural persons falls within the scope of freedom of expression, a non-paid promotional video by an influencer for a party or candidate can still be considered as an in-kind donation of a service (comparable to the free performance of an artist during an electoral rally). Ultimately, it is up to the Chilean oversight body to assess each instance on a case-by-case basis (Jaraquemada 2024). Conversely, in other countries influencers can take advantage of loopholes in regulatory frameworks. For example, in Brazil, while presenters on traditional media such as television or radio are not allowed to promote candidates, the rules do allow influencers to campaign on behalf of a candidate, as long as they are not paid to do so; paid digital campaigns are limited to candidates and parties running in the election (Grassi 2024).

In addition to influencers, the use of external experts or professional data and media companies to develop and carry out campaigns is also on the rise. This is the case in Nigeria, where such companies are often hired, but mostly work anonymously. Parties are sometimes supported with extensive and elaborate social media campaigns but claim that such efforts happen spontaneously—rather than in an organized (and financed) way on the initiative of the parties—as they claim not to have a budget for this (Adetula 2024). Without detailed financial reporting, it is difficult—if not almost impossible—to verify these claims. Evidence from the UK seems to indicate that expenses on digital consultants are massively underreported (Power 2024). In Chile, parties and candidates also frequently outsource digital advertising to external communication agencies, but do not feel bound by any accountability obligation, refusing to take responsibility for any statements disseminated by these agencies (Jaraquemada 2024). The engagement of external experts and marketing companies for online campaigning also appears to be a growing phenomenon in Kosovo. Yet, these expenses are not typically disclosed by the parties (Cakolli forthcoming 2025). In Albania, digital intermediaries such as consultants are also operating in the electoral space and are used by parties to obscure social media expenditure in their financial reports, leading to lower reported expenses than actually incurred (Power 2024).

Professional companies can also be used to imitate organic campaigning—the spread of unpaid political messages on social platforms—since this often falls outside the scope of campaign finance rules as it is considered a form of freedom of speech. However, while the political messages are not part of campaign expenses as such, parties and candidates can pay external companies or technology firms to artificially boost engagement with and the popularity of certain messages, relying on the algorithms of the online platforms to maximize their reach and visibility among voters. Indeed, such ‘bot strategies’ can be used to maximize the number of interactions to force the platform to deliver the content to more users (Grassi 2024).

One particular challenge in the use of digital marketing firms—particularly, the specialized data analysis service providers—is that they collect large amounts of citizens’ personal data with the specific aim of designing personally targeted campaigns to influence their voting behaviour. One of the most well-known examples is Cambridge Analytica in the UK. But it happens in many other countries, including outside the traditional English-speaking countries. In Nigeria, some political parties have been using digital service providers since 2013. For example, the All Progressives Congress party started working then with the American PR consultancy firm AKPD Message and Media—founded by senior members of Barack Obama’s 2008 presidential campaign—which specialized in quantitative research and data analysis to design political campaigns with the aim of influencing voter behaviour. Its main political competitor, the Peoples Democratic Party, responded by hiring the consultancy firm Cambridge Analytica in the run-up to the 2015 general elections (Adetula 2024; Akande 2021).

The use of digital technology platforms has been heavily criticized in the country, as it has been used to directly and indirectly attack opponents, spread false information and even incite violence. For example, the involvement of Cambridge Analytica by the Peoples Democratic Party was linked to the stealing of confidential material from the main opposing presidential candidate in an attempt to discredit him (Adetula 2024; Akande 2021; Kaiser 2019). In addition, there is a suspicion that the procurement of these external services might be partly financed through illegal funding and unlawful use of state resources (Adetula 2024; Akande 2021).

Some countries have made efforts to limit the potential impact of influencers or digital marketing firms on electoral campaigns. For example, in 2021, the Brazilian oversight body prohibited the hiring of individuals—influencers—or any other legal entities (e.g. digital marketing firms) to circulate paid advertising of ‘a political-electoral nature in the profiles, pages, channels—and more broadly digital environments—of the political candidates in electoral competitions’ (Brazil 2021). Yet, despite this ban, the presence of influencers during the 2018 and 2022 elections in Brazil was massive, and it remains difficult to monitor whether or not they received any financial remuneration for the electoral campaigning they did (Grassi 2024).

5.5. Coordination of oversight agencies

Because of the technical and legal complexities of online campaign activities, the task of monitoring these campaigns is often distributed across several oversight agencies. In order to achieve comprehensive oversight, a clear division of labour and effective collaborative procedures are essential. Such clarity is often not provided in the regulatory framework, and the responsibility for online political advertising lies at the intersection between the mandates of several different oversight bodies. Sometimes vague mandates can lead to some tasks not being exercised properly, or conversely, competition among agencies regarding their oversight responsibilities. Better synergies between all monitoring organizations involved are essential for successful oversight.

In Nigeria, control of the digital aspects of electoral campaigns is distributed between the multiple agencies, with the Independent National Electoral Commission at the centre of it. During the 2019 elections, there was an attempt to coordinate between the different actors involved. An Inter-Agency Campaign Finance Monitoring Group was established to promote synergies between the Independent National Electoral Commission and the other agencies involved—such as the Economic and Financial Crimes Commission, the Nigerian Financial Intelligence Unit, the Central Bank of Nigeria, the Nigerian Broadcasting Corporation, the National Communications Commission, the Centre for Social Justice, the Nigeria Police Force, the Federal Inland Revenue Service, the Independent Corrupt Practices and Other Related Offences Commission and the Nigerian Press Council—in monitoring the campaign income and expenditure of all parties and candidates. Yet, the coordination remained largely ineffective due to a lack of adequate support from the various organizations involved (Adetula 2024).

In Mexico, there are at least three bodies that oversee the elections: the National Electoral Institute (in particular, the Technical Audit Unit, which monitors online activities), the Federal Electoral Tribunal and the Special Prosecutor for Electoral Crimes. The complexity of online political campaigning requires these three institutions to collaborate strongly, but the rules—and consequently also the responsibilities for oversight—are often unclear, especially with regard to the use of influencers in political campaigns. Tackling possible misuse of political campaigning by such influencers requires substantial inter-agency coordination, which is not always easy to achieve in practice (Agrawal, Hamada and Fernández Gibaja 2021).

5.6. Fake news, disinformation and artificial intelligence

The use of digital technologies—and in particular social media—facilitates the spread of fake news and disinformation. For example, during the 2023 general elections in Nigeria, there was wide distribution of a letter claiming that one of the presidential candidates was involved in an investigation regarding drug-related accusations, even allegedly signed by a representative of the Independent National Electoral Commission. Although the letter was faked, it was widely shared on social media (Adetula 2024). In Bosnia and Herzegovina, parties and candidates rely heavily on false accounts and profiles on social media and websites to spread deliberate misinformation. Another strategy frequently used is the creation of so-called ‘portal farms’, which points to networks of websites and Facebook groups managed by a limited number of people, used to amplify the reach and content of social media posts, with a lot of misleading political content (Micanovic forthcoming 2025).

The latest threat, often linked to fake news and disinformation, is the emergence of the use of artificial intelligence (AI) in the creation of political advertisements. In particular, so-called ‘deepfakes’—computer-created manipulation of a person’s voice or likeness using machine learning to create content that appears real—pose new threats to the electoral process, as they might mislead voters by making them believe a politician made a certain statement or acted in a certain way, which did not happen in reality. There are clear indications that AI-generated content has already been used in the USA: for example, Ron DeSantis released a manipulated photo of Donald Trump in June 2023, and Trump and other candidates have also used deepfakes in their campaigns in 2024. Deepfakes also create new risks with regard to foreign interference. This is why the Federal Election Commission is considering a ban on deepfakes in campaign advertising. No regulatory action has been taken so far, but some platforms have taken proactive initiatives. Google, for example, announced that from November 2023, it requires political campaigns to label advertisements that utilize AI with disclaimers such as ‘this audio was computer-generated’ or ‘this image does not depict real events’ (Venslauskas 2024).

Based on the experiences with online campaigning in the examined case studies, a number of best practices and recommendations can be put forward.

6.1. Increase transparency

An important first step is to increase transparency around digital campaign expenditure. This allows for public scrutiny, fosters citizens’ trust and enhances political accountability (Agrawal, Hamada and Fernández Gibaja 2021). In this respect, regulators can look at the online platforms, the political parties and candidates, and the oversight bodies. Regulators can compel platforms to offer the highest degree of transparency regarding campaign expenditure. At an individual level, platforms can be obliged to provide the necessary background information for all published advertisements. At the collective level, platforms can be forced to set up public repositories that include all political advertisements. This allows citizens and civil society to trace the main trends in political advertising, and to compare the cost of the advertisements with the spending amounts as disclosed by the political parties and candidates, or the oversight bodies. The rules imposed by the EU can serve as a best practice, especially for the treatment of individual platforms.

However, while transparency notices and repositories for political advertisements offer a wealth of information on the external costs of digital campaigning (i.e. the costs of publishing—and possibly also targeting—campaign advertisements), the internal costs (such as personnel costs for communication staff, data scientists or campaign strategists, or the development costs of a political advertisement) are not included in such financial overviews. In addition, the expenses on consultants or other intermediaries who work for political parties and candidates should also be captured within such a transparency regime. A better overview of these aspects of digital campaigning can often only be derived from the financial reports drafted and published by the political parties and candidates themselves, or by the monitoring bodies. Therefore, it is important that regulators introduce a broad definition of online electoral expenditure, in combination with detailed reporting templates with a high degree of granularity. For example, such expenses should not be limited to the costs of publication of an advertisement on an online platform, but also include all costs related to the development of campaign content or the maintenance of social media accounts and websites, and the salaries/wages paid to communication or data management staff.

To facilitate oversight of these internal costs, the traceability of financial transactions is of the utmost importance, including the identity of the companies or individuals hired for the development and dissemination of political advertisements or messages. This is, for example, the case in Brazil, where the rules stipulate that all organizations or individuals that have been contracted for social media boosting campaigns should be identifiable through their corporate or individual taxpayer number (Grassi 2024).

Moreover, transparency and accountability measures should be applicable to all candidates and parties, regardless of whether they gain seats or not. Otherwise, candidates in non-electable seats can be used to circumvent certain campaign finance rules to benefit their political party overall. It can be more difficult for the digital aspects of campaigning to check this. Indeed, in some countries, such as Brazil, non-elected candidates are exempt from submitting financial reports about their electoral campaigns. In general, regulators and oversight bodies should enforce transparency in a way that allows citizens—and civil society more generally—to easily access the requisite information. Consequently, the development of user-friendly and accessible online repositories by social media companies is important (Wolfs 2022).

6.2. Include third parties in the regulatory framework

A second important element is to include third parties in the regulatory framework. Third-party campaigning is often omitted from campaign finance rules, and as a result it can be used to circumvent specific provisions and limitations regarding electoral expenses. Therefore, policymakers should adjust the regulatory framework in a way that it is also applicable to third parties—for example, by including all expenses from third parties on behalf of a candidate or party as actual candidate or party expenditure (possibly subject to a spending cap), as well as the requirement to officially register as a third party before an organization can engage in political campaigns. Ireland can provide best practice in this respect: the country obliges third parties to register with the oversight body, open a dedicated bank account for political campaigning and submit annual reports. Canada has similar rules in place. Latvia has introduced spending limits for third parties, obliges the use of disclaimers for online advertisements by third parties, and compels platforms to refuse the placement of third-party advertisements when the spending limits are exceeded (Power 2024). In any case, it is important to look for a proper balance between maintaining more control over third-party spending while not limiting this type of democratic engagement. In addition, it is important to extend the responsibility for the content of third-party campaigns to the parties and candidates they are campaigning for, so as to prevent political actors from escaping liability for possible infringements or disinformation.

6.3. Improve oversight over external media companies

Better oversight over external media companies can strengthen overall campaign monitoring. Oversight bodies can achieve this by setting up a public register of certified suppliers of external communication and PR services for online political campaigns. This is a measure that is being considered in Chile, for example (Jaraquemada 2024). It is believed that this could lead to better monitoring of the standardized services, better oversight of electoral expenses and fairer pricing, and thus prevent possible conflicts of interest, overpricing or abuse of public resources. Currently, it is often unclear for the oversight body which companies were hired by the parties and candidates, which services they have provided, and the prices for the development and implementation of the online campaign. In addition, such a public register of certified suppliers would also make it easier to extend the civil liability from these companies to parties and candidates, meaning that they could also be held responsible for any wrongdoing that might be established in the advertising services provided.

6.4. Clarify the differences between digital organic and funded political campaigns

A regulatory framework should clarify the differences between digital organic and funded political campaigns, and ideally create a level playing field between digital and physical electoral campaigns to make the regulatory framework as future-proof as possible. This also means that regulations should be generic and not platform-specific (Agrawal, Hamada and Fernández Gibaja 2021). Organic campaigning can be an important instrument for smaller or new parties and candidates with smaller campaign budgets (Agrawal, Hamada and Fernández Gibaja 2021), but rules should be clear about the limits and possibilities of such organic digital campaigns. If it remains unregulated, it could be misused by well-resourced parties to mimic a broad grassroots campaign.

If there is a specific electoral period, this should be applicable to both online and offline campaigning, and any prohibited periods for campaigning should apply to all types of electoral campaigns (including digital). Specifically with regard to instant messaging, oversight bodies should work with digital platforms to block coordinated mass messaging by individual accounts (i.e. reduce the number of messages that each individual account can send during an electoral period) and make it easier to identify the source and destination of massive electoral publicity. As already mentioned, this is an important campaign tool that has been used in Brazil to circumvent corporate spending on electoral campaigns (which is prohibited) (Grassi 2024).

6.5. Strengthen the overall capacity of oversight bodies

Another recommendation is the strengthening of the overall capacity of oversight bodies. There is a clear need for specific expertise to monitor the digital aspects of electoral campaigns, which requires additional financial and human resources. The monitoring bodies in most countries have limited resources. The Chilean oversight body, the Electoral Service, has a communication unit that specializes in monitoring the activity of parties and candidates on social media, although it suffers from a lack of capacity (Jaraquemada 2024). The oversight body in Kosovo, the Office for Registration, Certification and Financial Control of Political Entities, suffers from human and financial resource shortages, and limitations in technological capabilities and expertise. Also, the selected audit companies—responsible for auditing the political parties—struggle with a lack of personnel (Venslauskas 2024). An efficient way for oversight bodies to verify electoral expenses is the recruitment of a large number of observers that can verify whether political advertisements are reported or not, based on a sample. If they were not reported, they can be added to the total campaign expenses, which can lead to exceeding the expenditure ceiling (and thus potentially expose the party or candidate to possible penalties). However, this could also create an incentive for parties and candidates to fully disclose all electoral expenses.

The development of specific programmes to strengthen the capabilities of specialized knowledge and expertise within the oversight bodies—such as database management, data compilation and analysis, social media communication analysis, cybersecurity and information security, or financial transaction technology—is a good example of capacity building. In Latvia, for example, the oversight body relied on specific software to monitor sponsored political content online. Specific audits of the digital campaign activities of political parties and candidates should also be considered.

This exercise should be linked to improving the coordination between all involved oversight agencies, each with its unique capabilities and resources and specialized functions (such as data protection agencies, media monitoring agencies and so on). In this respect, there is the need for clarity in mandates, and proper coordination structures and procedures that have sufficient resources to implement the coordination processes effectively. In addition, the creation of regional and global networks, facilitated by international organizations and CSOs, to exchange ideas and best practices on the oversight of digital campaigning can further strengthen the overall oversight capacity.

6.6. Strengthen the capacity of CSOs

Similarly, it is important to strengthen the capacity of CSOs. They play a crucial role in monitoring because they assist the formal oversight bodies in their work to monitor campaign finance to ensure transparency and fair competition among all actors. In addition, they are often actively engaged in advocacy for political finance reform: in particular, as political parties set their own rules, there is often a need for external pressure from CSOs and the media to push them towards the necessary regulatory change. Yet, in many countries, CSOs lack sufficient resources or the specialized expertise to successfully take up these tasks.

Therefore, there is the need for capacity-building programmes on tracking and monitoring online campaign finance on the analysis of financial disclosure reports (for both the oversight bodies and the political parties and candidates themselves), and possibly on how to develop advocacy campaigns and civic education programmes to spread their message across the wider population (and to convince them of the detrimental effects for democracy of irregular party and campaign finance and expenditure).

The introduction of user-friendly and accessible tools for monitoring can also be helpful. A good example is the application developed by the Election Commission of India with the specific aim of helping citizens and CSOs report on possible campaign finance infringements. Similarly, it is important for oversight bodies to develop public databases—or to place the legal obligation on online platforms to develop them—containing comprehensive information on political advertisements, sources of funding and expenditure, in a user-friendly interface to help civil society in its monitoring efforts of campaign finance.

Samuel Power

7.1. Introduction

One can think of developments in online campaigning in one of two ways: as an evolution of or a revolution in existing practice. Indeed, political campaigns—at their heart—entail a process of communicating how effectively parties and campaigners have allotted (or will allot) tangible or imagined resources. Therefore, it is not at all surprising that we see campaigners moving with the times and using new technologies to get their messages heard. This was true for radio in the early 20th century and television a little later, through to the Internet and digital media at the turn of the millennium. Such is the nature of recent developments that academics now argue that we have entered a ‘fourth phase’ of political campaigns (Magin et al. 2017; Gibson 2020). This phase has been described as ‘data driven’ (Kefford et al. 2022), but it is generally characterized by a focus on the harvesting and forensic use of individual-level data to microtarget groups of voters (Votta et al. 2023). This means that the use of online and social media advertising is prominent, alongside large-scale data analysis and message testing.

Roemmele and Gibson (2020) have characterized these developments as either scientific or subversive. Scientific campaigns are those that interact with the new campaign environment in ways that mobilize and inform voters, whereas subversive ones focus instead on demobilization and fostering misinformation. The latter also feeds into wider concerns that recent online developments lead to echo chamber effects, whereby algorithmic sorting leads citizens to engage only in news stories structured around their own preferences. This argument is ongoing, however, and others suggest that platforms can provide more exposure to diverse viewpoints (Feezell 2018).

This argument on the benefits and perils of online campaigning brings us to the question of whether it leads to an equalization or normalization of campaign practice. Namely, scholars have asked whether the fact that digital technologies make campaigning cheaper and more efficient has benefited smaller, less established parties, and allowed them to compete on a more equal footing (Gibson and McAllister 2015). On the other hand, it is contended that larger, wealthier, and more established parties simply use online space to benefit from pre-existing inequities. In other words, have these developments led to a revolution in or an evolution of how campaigns function?

This latter issue is fundamentally interlinked with money in politics—and the resources political parties and campaigners have at their disposal. It is therefore useful to think about the national legal framework governing political finance in Albania before considering whether different areas are addressed by recent developments. The two key pieces of legislation are the Electoral Code of 2008 (as amended in 2012, 2015 and 2020) and the Political Parties Law of 2000 (as amended in 2006, 2011, 2014, 2017 and 2020), complemented by regulatory acts and decisions of the CEC.

7.1.1. Donations

Foreign donations are banned in Albania, although assistance that parties (or international unions of parties) receive from political foundations and organizations that have Albanian representatives is permitted. Corporate and trade union donations are permitted, but anonymous donations are prohibited. There is no limit on contributions made to political parties outside election periods, but contributions, including from a candidate’s personal wealth, made during elections may not exceed ALL 1 million (USD 10,550), which applies to both parties and candidates (and in-kind donations). The Organization for Security and Co-operation’s Office for Democratic Institutions and Human Rights (ODIHR) has recommended (2021; 2023) that reports detailing donations and expenditure be published prior to election day, and that existing limits on donations and spending be extended to include third-party campaign groups.

7.1.2. Spending

There are strict limits on the amount a political party or candidate can spend on an election. Political parties can spend no more than three times the amount the largest political party has received in public campaign funding. For individual candidates this figure is no more than 50 per cent of the largest campaign grant received by a political party. Third-party spending is not currently regulated. There are limits on spending on traditional media advertising during elections for political parties, and candidates, but these limits do not apply to social media advertising (only to advertising via ‘signal broadcasts’). Experts interviewed for this Report suggested that the consequence of this was that approximately 60 per cent of advertising during elections was effectively unregulated.

7.1.3. Public funding

According to article 88.1 of the Electoral Code, parties that attain over 1 per cent of the vote nationally during the most recent election of the same nature (general or local) are eligible to receive public funds to run an electoral campaign. The calculation of public funds received by the parties is based largely on the number of votes they gained during that election. Article 19 of the Law on Political Parties provides parties with annual public funds for their regular activities beyond electoral campaigns. More specifically, 70 per cent of the public funding is allocated to political parties based the number of members of parliament, 20 per cent is allocated equally to all parties that obtain over 10,000 votes and 10 per cent is given to parties based on the number of votes they received in the previous election. Parties also receive free—but not unlimited—access to state media.

According to article 15 of the Electoral Code, gender quotas also apply. This means that in every district at least one woman and one man must appear among the top three candidates on the party list, and each list must have at a minimum 30 per cent representation of each sex. Lists of city council candidates must include an equal number of candidates of each sex, and the lists must alternate between women and men. Failure to adhere to this requirement can result in sanctions of up to ALL 1 million and the subsequent refusal of the corresponding party list.

7.1.4. Disclosure

Political parties have to submit annual reports on their finances to the Central Election Commission (CEC), and are required to provide a report on their campaign finance within 60 days of the announcement of the election result (this 60-day period also applies to candidates). This report should be publicized no later than 30 days after its receipt or, in certain instances, after further verifications are completed. The identities of donors who donate more than ALL 50,000 (USD 525) are made public. In an effort to increase transparency, the CEC piloted a digital platform for political finance reporting and disclosure during the May 2023 local government elections. Likewise, ODIHR (2023) reiterated its recommendation to introduce periodic public disclosure of financial contributions to campaigns and expenditures before election day.

7.1.5. Oversight and sanctions

All reports are submitted to the CEC, which has the power to investigate wrongdoing and apply sanctions. Violations are punishable on a sliding scale ranging from ALL 50,000 to ALL 5 million (approximately USD 520 to USD 52,000), and in certain instances the forfeiture of illegal funds received and/or the suspension of public financing for up to five years.

7.1.6. Conclusion

In Albania there has been neither an evolution of nor a revolution in legislation given developments in campaign practice, and the law very much represents analogue solutions to a digital age. This does not mean, however, that the law needs to be rewritten wholesale. As was outlined at the beginning of this section, developments in the digital age of political campaigns do not necessarily need to be considered an existential challenge or threat. Often positive elements can be harnessed, and actors with malign intent can be effectively regulated. That said, regulation does need to move with the times, and as we will outline below there are significant gaps in the current legislative landscape and, moreover, examples of where the current rules, as written, are not enforced.

The need for evolution is clear. In numerous interviews with stakeholders, the possibility was raised that the law—if properly interpreted—could also apply to digital spending and digital campaign practice. However, a wider cultural issue was also highlighted, which is that unless legislation specifically outlines that an activity is to be regulated, it is unlikely to be so. Regulators have two approaches at their disposal: they can act as constructionists (i.e. work from what is written) or as activists (i.e. interpret electoral codes and treat them as living documents). Several of the people interviewed for this project said that the regulatory culture in Albania is one of strict constructionism. In other words, if legislation does not specifically require something, political actors are unlikely to actively address ongoing (and problematic) developments. Given this situation, the Electoral Code should be updated to reflect the way in which money is now spent in Albanian political campaigns.

This Report provides evidence of the use, trends and practices relating to online campaign spending through case studies of the 2021 parliamentary general election and the 2023 local government elections. Following these case studies are reflections on the challenges inherent in monitoring online campaign finance as well as on international standards and best practices. The Report ends with several suggestions for reform of the Albanian political finance regime.

7.2. Online campaign spending in Albania: A case study of the 2021 parliamentary general election

Table 7.1 contains a summary of data published by civil society on the election spending returns for the five parties with over 1 per cent of the total spend during the 2021 Albanian parliamentary election. As is, perhaps, unsurprising, spending is dominated by the Socialist Party (SP) and is considerably more fragmented beyond that.

The total party spending can be broken down by category. However, many of the categories contain so little spending as to be largely irrelevant to the overall picture of campaign spending in Albania. Therefore, Figure 7.1 simplifies the information, collating the spending of the five largest expenditure categories (or 87.1 per cent of the total reported spend). This ‘official story’ suggests that traditional media spending dominates over social media spending (though the distinction between these categories is somewhat unclear) as well as spending on other more conventional forms of campaigning, such as promotional materials. Moreover, wider campaign activities are utilized by the five main parties, which shows us that parties are using campaign techniques differently, and this takes us back to the issue covered in the Introduction concerning whether recent developments equalize or normalize political campaigns.

Here we see evidence of the growth of digital campaigning normalizing campaign practice in Albania in two ways. Firstly, we see that the SP and the Democratic Party (DP) spend significantly more than their rivals on social media, and that they are the only parties to use campaign consultants (and to spend a large proportion of their campaign budgets doing so). Secondly, the Socialist Movement for Integration spends considerably more than the SP and the DP on ‘promotional material’, which could broadly be understood as more traditional campaigning based around leaflets and posters. Therefore, initial evidence on spending dynamics indicates that the move to online campaigning is serving to benefit wealthier parties rather than to raise the platform of smaller or newer parties.

7.2.1. Campaigning on Facebook

The official story likely provides only one part of the picture of the way in which online spending occurs in Albania, and the Facebook (now Meta) advertising archive can provide a fuller view. Conducting analysis utilizing the Meta advertising archive is imperfect (see 7.2.5: Challenges in monitoring and oversight of online campaign finance), and it is still hard to determine precisely how much is spent on this platform during and between elections. That said, we can ascertain some basic facts. The ad archive has been active in Albania since 11 March 2021, and since that date 27,094 adverts have been placed at a cost of ALL 65,279,933 (USD 689,000); these ads are broadly categorized as relating to ‘social issues, elections or politics’.² Table 7.2 provides a simple descriptive breakdown of the 18 biggest spenders on the ad archive between 11 March 2021 and 31 August 2023. This breakdown captures overall expenditure during this period (including normal activity during non-election years); that said, much of the activity occurred during election periods in 2021 and 2023. Due to the challenges of collecting historical data from the Facebook advertising archive, it is not possible to say with certainty how much of the information in Table 7.2 includes spending in the respective election periods. If we look at each page in Table 7.2, however, we see that the vast majority of adverts are placed during regulated periods.

While aggregating the highest ad spenders on Facebook is quite a rudimentary analysis, even this snapshot provides some interesting information, and shows weaknesses in the Albanian political financing regime. If we continue to use the 2021 general election as a case study but consider other data sources and real-time election monitoring by organizations such as Balkan Investigative Reporting Network (BIRN) Albania, we can make a number of observations, as outlined in the following sections.

7.2.2. Significant (and unregulated) third-party activity during the election period

Among the top 18 spenders in the Facebook advertising archive are two third parties—Untold Stories of Albania and Punë Punë. These pages—both pro-SP (Likmeta 2021)—have not been active since the election but collectively spent ALL 2,263,635 (USD 23,900) during the election. That these pages remain in the top 18 all-time spenders on Facebook ads in Albania, over two years after the election, gives some insight into the relative level of third-party spending.

Analysis by the Coalition for Reforms, Integration and Consolidated Institutions (KRIIK) (2022) highlights three more third parties of note during the 2021 parliamentary election: Te Jesh Grua (ALL 537,059, or USD 5,670), Pakëz Dritë (ALL 478,456, or USD 5,050) and Ligj dhe Rend (ALL 414,635, or USD 4,375). The spending by these five organizations (all of which are sympathetic to the SP) amounted to a total of ALL 3,693,785 (USD 39,000), which, if the official returns were accurate, would easily amount to the third-largest social media spend by any political party during the election. In fact, if we return to the official story of election spending returns, we can see that the total social media spend was reported as ALL 21,266,999 (USD 224,500). Therefore, spending by these five third parties represents 17.4 per cent of total social media expenditure during the 2021 parliamentary elections. However, as we address below, the official story contained in election spending returns is likely an underestimate. Wider analysis during the 2021 election estimated the true social media spend—from 11 March to the eve of the election—as being approximately ALL 28 million (USD 295,500). That these organizations also support the SP suggests that this lack of regulation of third-party spending masks the true scale of normalization that is at play in Albania. The SP’s spending on social media far outweighs that of other parties anyway, and large amounts of support from third parties only exacerbate such inequities.

7.2.3. Under-reporting of social media spend

There is good evidence that neither official spending returns nor the Meta advertising archives provide a full picture of spending on social media. We can see an example of this if we simply look at the top 18 spenders in the ad archive. Number 16 in Table 7.2 is Astrit Patozi, a candidate for the Party of Democratic Conviction (PDC). During the campaign he spent ALL 695,005 (USD 7,330) on 67 adverts, yet the official CEC returns state that the PDC spent nothing on social media. Similarly, we can see that Fatmir Mediu, a candidate for the Republican Party of Albania (RPA), spent ALL 335,962 (USD 3,540) during the 2021 campaign on their personal Facebook page, even though the RPA’s total reported social media spend was also zero (Open Data Albania 2022).

The above-mentioned findings are echoed in reporting from KRIIK (2022), which states that during the 2021 parliamentary election ‘33.2 million ALL was spent on political ads on Facebook and Instagram’, while ‘a total of about 28.4 million ALL’ in expenditures was declared on official spending returns. Moreover, as KRIIK notes, this analysis is merely of spending on Facebook. Given that other social media platforms are in use during elections, the true spend on social media is likely much more (see also Box 7.1). Unfortunately, Google does not provide a political advertising archive for Albania. We know from recent research that political parties are strategic in the platforms that they choose to spend money on. For example, in the United Kingdom the Conservatives spent as much on Google and YouTube banner ads as they did on Facebook (see Dommett et al. forthcoming; Dommett and Power 2022). Therefore, we know that it is likely that neither official spending returns nor the Facebook ad archive provides a full picture of spending. And we also know that, even if they did, parties are using other platforms to ensure that their campaign communications are heard, and a true analysis of spending on these other platforms is not yet available.

7.2.4. Digital intermediaries

Finally, there is also evidence of digital intermediaries operating in the electoral space in Albania, and further cementing the hold of the wealthier, established parties in this space. Of lesser concern than the more blatant use of social media to circumvent Albanian election law, the use of digital intermediaries remains another way in which social media expenditure can be obscured on official election spending returns, thus highlighting the fact that the current picture of social media use may not be entirely accurate. Three candidates—Alqi Bllako, Eduard Shalsi and Dallendyshe Bici (all from the SP)—had adverts placed by WOW Digital, which expenses for these ads. Total spending amounted to ALL 919,282 (USD 9,700), which can reasonably be—and from evidence from other cases and context is likely to be (see Dommett and Power 2023)—reported as something other than social media expenditure on official CEC election returns (e.g. consultancy or campaign materials).

7.2.5. Challenges in monitoring and oversight of online campaign finance

We conducted eight interviews with stakeholders in the electoral process in Albania, including CSOs, practitioners and international experts. These interviews highlighted a number of issues with the current political finance regime, and in particular with the monitoring, regulation and enforcement of online campaign financing. Firstly, it was noted that, while there are strict spending limits for political parties and candidates, there was a lack of regulation on these expenditure ceilings on online campaigning.

Moreover, the law does not properly recognize that politics continues beyond elections. While the Electoral Code (article 92.2) suggests that limits apply to any campaign expenditure, regardless of when it is made, in reality the provision for ‘electoral campaign purposes’ is vague. Political parties and candidates seek election; therefore, any spending can be reasonably considered to be for electoral purposes. The law should, therefore, be clearer on what an electoral purpose is, or expansion of the regulated election period should be considered.

Similarly, a lack of regulation of third parties is increasingly problematic. There is clear evidence of sizeable third-party spending on Facebook/Meta (and potentially on other platforms that do not disclose data on such spending) during campaigns, and we also know that third parties can have significant influence between elections—and often operate primarily outside of election periods. Instead of a ban, which is largely unenforced on digital platforms, it is better to limit third-party spending or to adopt regulations on third parties in line with regulations on parties so as to (a) have more control over third-party spending; and (b) not limit the genuine benefits of this kind of democratic engagement.

Comparatively speaking (see Chapter 2), the disclosure requirements are—at least in theory—adequate. There are clear requirements for itemization, and online campaign spending is specified on returns that can be accessed. However, we know from existing research that for transparency to be an effective means to promote accountability and increase trust, it needs to be implemented effectively (Alessandro et al. 2021; Power et al. 2023). In other words, spending transparency needs to be enforced in such a way that citizens can easily access the requisite information, which has not been the case in Albania for decades.

This study relies primarily on data provided by the CEC, secondary sources (such as reporting by KRIIK, BIRN and Open Data Albania) and our interviews. Meanwhile, important steps have been taken—including with input from the International Institute for Democracy and Electoral Assistance (International IDEA) (see Wolfs 2022)—to develop and introduce a digitalized disclosure system, and efforts should continue to ensure its successful implementation. Without this a transparency regime risks being transparent in name only and unlikely to serve its stated aims of (a) reducing corruption; and (b) increasing trust. What is important is to ensure that the transparency regime provides complete and consistent returns, and that these returns are accurate and accessible (Dommett and Power 2023).

The above evidence shows that it is likely that online campaign spending in Albania is under-reported through official means. This issue is not unique to Albania: in almost every country it has been incredibly hard to gain a true picture of the breadth of spending online (Dommett, Kefford and Kruschinski. 2023). The official story is often drawn from publicly available (and legally required) party and candidate accounts and financial returns. These returns, more often than not, provide information only where there is direct spending on online platforms (such as Facebook, TikTok or Instagram). Spending on campaign consultants, however, who might well place advertisements for political parties and candidates as part of the wider suite of the services that they provide (as we have seen in the Albanian case), is not included. Therefore, a significant piece of the picture is missing.

Information in advertising archives is not always made public by online platforms, and when it is, it is often imperfect or inexhaustive. Among platforms that proactively disclose information, Facebook/Meta, for example, provides only banded estimates of spend, and there are numerous other challenges with regard to analysing the data they provide (see Dommett and Power 2019). There are also problems relating to spending on online platforms—specifically with regard to spending limits. Due to the nature of virality and of organic versus paid content, as well as the use of influencers (for more on these issues, see Riedl, Lukito and Woolley 2023), different adverts can cost vastly differing amounts—and prices can rise and fall (see Dommett and Power 2019; Nadler, Crain and Donovan 2018). As online campaigning evolves, issues arise not only with campaign spending itself but also with the way in which platforms operate and how users interact with them.

This dynamic can mean that the true nature of campaign spending is known only after the campaign has ended (allowing political parties and candidates to exceed spending limits, either by accident or by design). Moreover, ‘these ambiguities mean that even if greater detail on advertising spending was provided, it would remain difficult for observers to determine the extent to which paid advertising intersects with organic reach’ (Dommett and Power 2019: 264). Finally, countries find themselves subject to the whims of the online platforms themselves, which have different priorities from those invested in creating (and maintaining) working democracies (though the EU is leading the charge when it comes to regulating platforms). This situation makes it incredibly hard to compel information from the platforms and leaves significant space for progress on platform transparency to simply be reversed.

Another perennial problem for all those interested in better regulation of online spending, articulated by many participants of this project in Albania, is how fast-moving the space is. Regulation needs to keep up with evolving platform practices (and priorities), and parties (and campaigners) use different tactics (and different platforms) from election to election. The slow pace of regulatory changes against the rapidly changing environment can leave many feeling that elections online are reminiscent of a political Wild West (Palese and Mortimer 2019), a space with few rules and awash with malign actors, or that EMBs are simply ‘analogue regulator[s] in a digital age’ (Kinnock 2019: 27). This problem is exacerbated by the slow pace of legislative change: as one challenge is mediated and addressed, another may appear. In the wider (academic) world of campaign finance this is known as the hydraulic theory of money in politics (Issacharoff and Karlan 1999). In other words, money in politics is like water: it flows uncontrollably and will find the weakness in the dam through which to break. No legislation is perfect, so there will always be cracks (i.e. loopholes), and it is here that money will flow. At present, in many countries (Albania included) the space where these cracks are to be found is online.

Technological change will always outpace regulation, and this lack of first-mover advantage is compounded by an inherent (and understandable) conservatism within electoral commissions themselves. Unlike other regulators, they are in a uniquely political position, consistently at risk of interference and threats (overt or otherwise) from those in power (see Cheeseman and Elklit 2020), which often means that they do not wish to be seen to be making directly political decisions. While this is a broad truth regarding electoral commissions the world over, it was also a concern specifically voiced by some with regard to the CEC in Albania.

In our interviews, a challenging dynamic was presented. A common recommendation was to require more granular information on spending returns in terms of a wider range of expenditure categories (which could capture the range of online spending). Given the pace of change in this space, however, it could well be the case that as soon as these categories are given legislative footing they are no longer relevant. For example, as the ongoing debate over the extent to which organic content promoted by influencers should be covered by regulation continues, it will be important for the CEC to have built-in flexibility in its competencies. Some interviewees said such flexibility would create a form of constructive ambiguity that empowers the CEC both to better investigate and to move faster than the legislative process allows.

Our interviewees agreed, in general, that the regulatory challenge in Albania is not an existential one, but something faced by electoral commissions (and countries) the world over: these agencies simply lack the resources and technical capacity to keep up with online developments. Albania’s CEC operates on a limited budget, with limited personnel. Many countries face similar challenges, and regulatory gaps are often addressed with an active, functioning and robust civil society that can provide extra-legislative support. However, interviewees from civil society agreed that, while there are a number of strong CSOs that provide such support, they are almost entirely donor-led. Therefore, if donors decide to remove their support, this important work may not continue. Currently, however, the accountability ecosystem—made up of a vocal and articulate civil society with a well-functioning electoral commission—is strong. This observation suggests that reforms should focus on strengthening existing legislation and empowering existing institutional structures, as opposed to pursuing more fundamental reforms (at least in the short term).

These issues aside, an increasing number of countries have grasped the nettle and attempted in various ways to address the challenge of online campaigning. The next section of this chapter summarizes information from International IDEA’s Political Finance Database, before focusing on specific examples of regulatory interventions and areas of best practice.

7.3. Thinking comparatively: Experiences and practices from other countries

The Political Finance Database provides the most detailed comparative summary available and highlights the general challenge of regulating in this area. Of the countries for which data is currently available, 82.3 per cent have no limits on online media advertising spending in relation to elections, 8.9 per cent have limits for candidates, 7.0 per cent have limits for political parties, and 1.9 per cent have limits for third parties (International IDEA n.d.). The most regulated region is the Americas, but, given the general trend of a lack of regulation in this space, the story remains one in which specific online spending limits are rare (see Table 7.3).

This dynamic of lack of regulation of online spending is, however, largely in keeping with the reluctance to place limits on advertising spending in general during elections. If we consider this data, we see a less pronounced but definite preference for no limits (see Table 7.4).

Limits are not the whole story, however, and if we consider whether any other restrictions on online media advertising exist, we see that 21 of the countries (or 15 per cent) for which we have data have some form of restriction. The type of regulations in place varies widely, ranging from the requirement to register online advertising profiles (e.g. Argentina and Indonesia); outright bans (e.g. Comoros); targeted bans, such as on election day and the eve of polling (e.g. Armenia and Malaysia); or bans on payments to influencers (e.g. Brazil); to regulations on specific content appearing in online adverts (e.g. Iraq, Paraguay, Philippines).

It should also be recognized that many countries are subject to effective proxy bans on online advertising, as their regulation on advertising includes periods of silence. However, we heard from multiple stakeholders in the electoral process that in Albania the period of silence is interpreted as concerning traditional media only and is largely ignored in the online sphere. Evidence from the Facebook advertising archive bears this conclusion out. There were multiple examples of political parties and candidates having adverts in place (or still viewable) on polling day during the last general election.

A final consideration is indirect online financing regulation, through requirements to itemize spending returns, which is much more common among countries across the world. For those countries for which there is data, 66.5 per cent do legislate that reports from political parties and/or candidates must itemize expenditure. This finding contrasts with the 26.2 per cent that do not, and the 7.3 per cent to which this question does not apply (see Table 7.5). The level to which spending must be itemized varies, however, and in some countries this form of regulation provides little information about the level of online spending. It is useful to think of these returns as having a high, medium or low level of itemization, which provides the researcher with high, medium or low levels of actionable information about types of election expenditure.

The next sections outline the different levels of itemization seen in the UK (high), New Zealand (medium) and Australia (low). More granular detail (when presented effectively) should be the aim in a functionally transparent political finance regime (see Dommett and Power 2023). The cases below are presented as a range of adoption practices—from low to high—that cover a suite of options. While it would be preferable to adopt the most complete and consistent model, it may also be the case that lesser models are more realistic in this instance—and at this stage.

7.3.1. High-level itemization: The case of the UK

The UK is considered by political finance experts to have a transparency regime that is effectively world-leading, at least in terms of the information provided (see Power 2020; Dommett and Power 2023). Political parties, candidates and third parties must report on expenditure during election periods, and they must include PDF invoices for any spend over GBP 200 (USD 250). Currently, the information provided by parties (the rules differ slightly for candidates and third parties) must be classified under one of nine headings: advertising; campaign broadcasts; manifesto or referendum material; market research/canvassing; media; overheads and general administration; rallies and other events; transport; unsolicited material to electors (see Figure 7.2).

However, there are still flaws with this system as it relates to the transparency of online campaign financing. Most notably, there is no category that specifies either online campaigning or online advertising (as there is in Albania, for example, as well as in countries such as Austria and France). This situation means that one has to rely on keyword searches for specific providers (such as Facebook, Google, TikTok, etc.) to gain a picture of spending on these platforms. Moreover, this kind of itemization provides information only on instances of direct relationships between online campaign spending and service providers. It does not provide information on the ‘digital intermediaries that have emerged in the electoral marketplace’, who ‘are regularly working with political parties to curate political adverts on platforms’ (Dommett and Power 2019: 262).

7.3.2. Medium-level itemization: The case of New Zealand

An example of medium-level itemization can be found in New Zealand. Here, political parties are required to provide line-item disclosure on all election expenditure, but, unlike in the UK, parties are not required to provide an initial (even broad) categorization of the type of expenditure. Primary expenses are instead carried under the headings ‘Party advertisements solely promoting the party’ and ‘Party advertisements shared with candidates or other parties’. As Figure 7.3. shows, however, the types of spending that occur under these broad headings can include anything from Facebook advertising, posters and music licensing to payment for a sign language interpreter. Moreover, there is little further detail (in the form of specific invoicing returns) on the manner of digital advertising provided by, for example, Together Communications, which could be performing a whole range of online campaigning-related activities. This means that it is even more difficult for the average citizen, or seasoned researcher, to calculate the true total of online campaign spending in this context—beyond very broad (and likely insufficient) estimates.

7.3.3. Low-level itemization: The case of Australia

In Australia we can see an example of low-level itemization. Requirements for itemization are built primarily around donations, as opposed to expenditure: most notably, political parties are not required to provide election spending returns. This means that, while detailed donation itemization can be found on candidate election returns, very little information on spending is itemized. An indicative example from the general election of 2022 can be found if we look at the PDF return submitted by Allegra May Spender, who, according to the Australian Electoral Commission Transparency Register (see Figure 7.4), spent more than any other candidate during the 2022 general election. The register provides significant details on donors (including amounts donated, the date of donations and the donor’s address), but significantly less on spending—simply that the candidate did indeed spend AUD 2.1 million (USD 1.4 million).

7.3.4. Best practice examples of online spending regulation: Ireland, Latvia and Canada

There are, then, lots of ways to better regulate online campaign spending either directly or indirectly. However, there are also countries that represent specific areas of best practice that are worth highlighting. One of the most detailed recent examples is Ireland’s Electoral Reform Act (Republic of Ireland 2022), which—alongside a raft of wider reforms—introduced significant regulation on online political advertising, including the following:

• digital imprints (including the text ‘political advert’ prominently displayed, a button/hyperlink to a ‘transparency notice’ that includes details on the buyer of the ad, confirmation of whether microtargeting was used, confirmation of whether lookalike targeting was used, the total amount paid or due to be paid, the estimated size of audience, the number of views, the relevant election or referendum, a link to the platform ad archive) (article 121(1), (2));
• obligations on online platforms to verify the identity of purchasers of online political adverts (article 122(1));
• strict limits on adverts commissioned from abroad unless they are commissioned by (a) a citizen of Ireland; (b) a company, corporation or unincorporated body that maintains offices in Ireland; or (c) a citizen (or company, corporation, unincorporated body or political party) of an EU member state and the advertisement relates to elections to the European Parliament (article 125(1));
• empowerment of the Electoral Commission to (a) monitor online political advertising; (b) monitor online platforms; (c) monitor buyers of online political advertising; (d) educate, investigate and enforce compliance (and non-compliance) (articles 127(1);
• clear regulation of third parties that states that they must (a) register with the Standards Commission; (b) open a bank account for political donations and (c) submit annual returns for political donations; and
• strong sanctions that range from fines to imprisonment of up to one year (in the case of summary conviction) or up to five years (in the case of conviction on indictment) (article 140(1)).

As the above-mentioned regulations have not yet been fully implemented, their effects are yet to be fully understood. Latvia, on the other hand, represents an example of a slightly more established best practice in this area, covering income, expenditure and reporting (see Cigane 2022). In terms of online campaign spending, regulation can largely be found in the Law on Financing of Political Organizations (Parties) and Pre-Campaign Election Law including the following:

• detailed itemization of spending, including a category for ‘[advertising placed] on the internet, except for the website of the political organization (party)’ (Law on Financing of Political Organizations (Parties): section 8.2); and
• regulation on third parties, including (a) spending limits; (b) inclusion of disclaimers on adverts placed online by third parties; and (c) an obligation on the part of platforms to refuse to place third-party adverts if they break spending limits (Pre-Election Campaign Law: section 5).

Canada also provides an interesting best practice, having updated the Canada Elections Act 2000 (Government of Canada 2000) in 2018 and most recently in 2023 to define online platforms and impose obligations on them concerning digital ad registries (i.e. advertising archives). The Act provides specific guidance for online platforms, which includes the following:

• a clear definition of what an online platform is: ‘an Internet site or Internet application whose owner or operator, in the course of their commercial activities, sells, directly or indirectly, advertising space on the site or application to persons or groups’ (article 319);
• required visitor thresholds for advertising repositories of online platforms: (a) in English (3 million unique visitors in Canada a month); (b) in French (1 million unique visitors in Canada a month); and (c) in another language (100,000 unique visitors in Canada a month) (article 321.1(1));
• guidance on two kinds of adverts, ‘partisan’ and ‘election’, which need to be included in ad registries: (a) partisan adverts are ‘ads that promote or oppose a party or candidate’; and (b) election ads encompass the definition of partisan ads with the addition that they also include ‘advertising that takes a position on anything that is or may become an issue during a federal election campaign’ (articles 2.1, 325.1(2));
• guidance on the information that should be included in the registry: (a) an electronic copy of adverts; and (b) the name of the person that authorized the advert in question (article 325.1(3));
• penalties, including fines or imprisonment, for non-compliance (article 495.1(a.1)); and
• a requirement that these regulations apply to third parties as well as registered political parties (articles 325.1(2), 349.6(1), 353(1)).

Canada also has an effectively world-leading system of third-party regulation, in terms of both online and offline campaign activity. The regulations in question include the following:

• a clear definition of a what a third party is: ‘a person or a group, other than a registered party, registered association, candidate or nomination contestant’ (article 348.01);
• guidance on who can register as a third party: (a) an individual who is a Canadian citizen or permanent resident; (b) a corporation carrying on business in and incorporated in Canada; (c) a corporation carrying on business in Canada but incorporated outside Canada, as long as the primary purpose is not to influence Canadian voters; or (d) a group (where the person responsible for the group is a Canadian citizen) (article 349.6(2));
• guidance on when third-party registration is required: (a) immediately after regulated activity is conducted in an election period that totals expenses of CAD 500 (USD 375) or more (article 349.6(1));
• a public database, the Registry of Third Parties, hosted on the Elections Canada website (article 358.2);
• the requirement for a separate bank account for the sole purpose of the campaign in question, in line with prior legislation set out in the Bank Act (article 358.1(1));
• guidance on three types of regulated activity—election advertising, partisan activities and election surveys—that includes (a) election advertising (‘an advertising message that promotes or opposes a registered political party or candidate, including by taking a position on an issue with which the registered political party or person is clearly associated’; (b) partisan activities ‘that promote or oppose a political party, nomination contestant, potential candidate, candidate or party leader, other than by taking a position on an issue with which the political party or person is associated’; and (c) election surveys ‘about voting intentions or choices, or about an issue with which a registered political party or candidate is associated, that a third party conducts or causes to be conducted … [where the] results are used in deciding whether to organise and carry out regulated activities’ (article 349);
• spending limits, adjusted each year for inflation, for third-party activity (articles 350 (1)(2)(3)(4)(5)); and
• campaign reporting requirements detailing all inflows and outflows no later than four months after an election (article 359.1).

7.4. Considerations for strengthening the online campaign finance regime in Albania

The Report began with a suggestion that countries can take one of two approaches: to enact a legislative revolution or to simply evolve alongside solid legislative principles. The preferable option in the Albanian case is to evolve current regulatory practice and, more importantly, to establish mechanisms and support systems such that pre-existing legislation can be implemented effectively. Moreover, developing more stringent regulation could also help to push back against the evidence that indicates that online spending is helping more established and wealthier parties to further consolidate their positions.

7.4.1. Enacting current regulatory principles

Throughout this Report evidence has been presented that either the current campaign finance rules are not being adhered to or campaign spending remains unregulated in the online space. This observation is based on misreporting of social media spending, non-adherence to periods of silence and unregulated third-party campaigning in support of established political parties (namely the SP). Taken together, while there is no concrete evidence that political parties have broken the rules surrounding expenditure limits, if campaign spending is not being accurately reported we have no good evidence that these expenditure limits are in any meaningful way effective.

The most important reform for the Albanian political finance regime is, therefore, quite simple. The CEC should be empowered through further resourcing to better fulfil its functions as an electoral regulator, such that it can:

• educate political parties and campaigners on the current political financing rules and correct means of reporting;
• investigate wrongdoing in a timely manner such that online adverts in contravention of the rules can be removed in (near) real time; and
• sanction political parties, candidates and third-party organizations that break these rules to form an effective deterrent against these activities.

Moreover, on paper the transparency regime is broadly aligned with international standards, and requirements for the itemization of electoral expenditure are at a higher level when we look comparatively. However, transparency is functional only if citizens easily access information and the information is easily available. Ongoing efforts on the part of the CEC to pilot and consolidate a digitalized political finance reporting and disclosure system are moving in this direction.

The current system in Albania relies on a patchwork accountability ecosystem of an official electoral regulator that is supported informally by a vibrant civil society, and excellent work is conducted by organizations such as BIRN, Equality in Decision Making Women’s Network, KRIIK, Faktoje, and the National Democratic Institute (NDI). These local organizations and international partners can effectively campaign for reform and highlight non-adherence to electoral law. That said, these organizations are largely donor-funded, which means that the accountability ecosystem is fragile. This fragility makes it even more important that the CEC be well supported and empowered in the event that international donor funding is no longer available.

7.4.2. Evolving current legislative practice

While much of the challenge that online campaigning presents can be met by current legislation and by better supporting the CEC, the following reforms should be made, which would serve to evolve current regulatory principles:

• renew efforts to bolster the capacity and effectiveness of the CEC;
• make party accounts easily available and accessible on the CEC website;
• ensure that more granular reporting on online spending is included in transparency returns;
• put a digital imprint regime into effect;
• ensure that third-party activity is more tightly regulated;
• place links between election expenditure returns and online platform advertising archives;
• require that platforms and the CEC uphold these rules; and
• where not covered in the above recommendations, harmonize regulations with incoming EU legislation and the best practices of EU member states.

7.4.3. Winning elections the right way

Many of the above recommendations echo those made—at the more general level—in International IDEA’s Winning Elections the Right Way (Heinmaa 2023). Based on a survey of EMBs from 28 European countries and 9 interviews, the Report outlined the numerous ways in which the regulatory challenges presented by the rise in online political advertising should be addressed. Firstly, it suggests four broad ways in which online political advertising is currently being tackled:

• no regulation of either online or offline advertising/media;
• regulation of some forms of offline media, but not online media;
• generic offline regulation that applies by proxy online; and
• explicitly regulated.

Albania provides a rather weak example of the third type of regulation. Although there is no specific online regulation, we do see some forms of online campaign activity captured in the current regulation, such as periods of silence (although we also find that this regulation is not adhered to). This is currently insufficient—not least to match EU accession requirements. So, mirroring the template set out in Heinmaa (2023), legislation in Albania should do the following:

• define what constitutes a political advert online and determine how new legislation needs to be drafted to reflect this definition;
• establish common standards for identifying political adverts—in collaboration with parties, social media platforms and experts—and identify who purchased them;
• create clear expectations for transparency in the online space;
• establish responsibilities for platforms to create legal representatives, transfer information/data to specific electoral stakeholders, communicate policies for moderation, and adjudicate complaints and assess risks; and
• clarify the extent to which the content of ads will be banned and the degree to which the regulators will be responsible.

Alongside the legislative reform, a number of soft-law or soft-power mechanisms can be used, in the form of voluntary codes of conduct that establish voluntary behavioural standards. While these are often criticized for being toothless (e.g. Hough 2017: 108–10), there is good evidence—not necessarily relating to political campaigning—that they can be effective. An example of a soft-power mechanism is the Extractive Industries Transparency Initiative, whereby reformers signalled their good intentions, but international actors then rewarded achievements such that eventually a wider norm of better practice was diffused (David-Barrett and Okamura 2016). Of course, establishing common standards and practices can be hard, and voluntary codes of conduct obviously lack a true accountability/sanctioning function, but as an initial step they can be useful, and can often form the parameters of a more institutionalized agreement.

An additional recommendation for the shape that a voluntary code of conduct could take is based on the Dutch model (Heinmaa 2023: 57), which is outlined in Table 7.6.

This Report also supports two more findings from Winning Elections the Right Way (Heinmaa 2023):

• the incorporation of flexibility into regulatory and oversight structures; and
• increasing resourcing and training for EMBs to meet the evolving challenge.

7.5. Conclusion

On paper, Albania has strong regulatory principles that should help to address the challenge that online campaigning, and online campaign spending, presents. In practice, however, these principles are not always effectively enacted. The challenge may not be existential, and it does not require a wholesale legislative revolution. A clear empowerment of the CEC and legislative reforms in the model of international best practice ought to be sufficient to manage developments concerning online political finance practices. However, the effort needs to be clear and sustained over time. These are challenges that many consolidated and long-standing liberal democracies struggle to keep pace with. So, absent a diligent approach, the threats are likely to grow into something altogether more concerning.

Gunther Vanden Eynde³

8.1. Legal framework: What we do not understand, we fear

In many European democracies, laws limiting campaign spending or prohibiting the use of a number of campaign instruments are a common practice (Gibson 2020: 51 and 215; Lilleker, Tenscher and Štětka 2015: 753). This is also the case in Belgium, where electoral campaign regulations fall under a federal mandate regardless of the nature of the election. The Federal Act of 4 July 1989 on the Limitation and Oversight of Election Expenses Incurred for the Election of the Federal Chambers as well as on the Financing and Open Accounting of Political Parties regulates federal elections. European and regional elections are both regulated by a separate federal act of 19 May 1994. Local elections, on the other hand, fall under a regional mandate. During the official campaign period for all elections (local, regional, federal and European), a strict spending cap limits campaign spending. Furthermore, advertisements on both radio and television are banned, as is the distribution of small gifts, the use of commercial billboards and the conduct of commercial campaigns over the phone (Federal Act of 1989: articles 2 and 5; Federal Act of 19 May 1994: articles 2 and 5).

In 2007 online social networks and digital ads were still far from mainstream. Even the United States was still a year removed from the Obama campaign of 2008, which is generally seen as a benchmark when it comes to the successful use of social media in electoral campaigns (Bimber 2014; Spierings and Jacobs 2014). It is in this context that the Belgian legislature decided to amend the above-mentioned federal act of 1989, adding online ads to the list of prohibited campaign instruments during the official campaign period for all elections (at the time, three months prior to election day). However, audiovisual content (even when made by commercial providers) on candidates’ personal websites and party websites was allowed (Chambre des Représentants de Belgique 2007a: 5; 2007b: article 5). This amendment was rushed through parliament to make sure that it would enter into force by the beginning of the official campaign period for the 2007 federal parliamentary elections, which started on 24 March (Chambre des Représentants de Belgique 2007a: 3). During this official campaign period, restrictions apply regarding campaign spending and the use of certain campaign tools (Federal Act of 1989: article 5).⁴ The rationale for this amendment, as mentioned in the parliamentary proceedings, was the great number of unknowns regarding the use and influence of this specific online campaign tool (Chambre des Représentants de Belgique 2007a). Due to these restrictions on the use of digital ads, the digital toolbox for parties and individual candidates was limited to websites and emails. Again, the only exception was for local elections. Although online ads have never been banned at the local level, they have not been very popular at that level so far (see e.g. Vanden Eynde et al. 2019).

In 2013, however, the Federal Act of 4 July 1989 was amended again, and the restriction regarding online ads was removed by article 9 of a federal act of 22 November 2013 (Chambre des Représentants de Belgique 2013b). In contrast to the prior amendment, which was rushed through parliament in order to be in force for upcoming elections, the 2013 amendment was not applied to the 2014 concurrent elections and entered into force only on 1 January 2015 (Chambre des Représentants de Belgique 2013b: article 37). The rationale mentioned in the parliamentary proceedings read as follows: ‘The difference (from offline political ads) can no longer be justified in today’s technological context’⁵ (Chambre des Représentants de Belgique 2013a: 8). These restrictions have also been lifted in legislation concerning both European and regional elections. Consequently, the official campaign period for the concurrent elections in 2019, which started four months prior to election day, was the first in which both political parties and candidates were allowed to use (targeted) online political ads as part of their campaign for European, federal or regional parliamentary elections. Whether this new tool found itself immediately on an equal footing with more traditional campaign instruments is the focus of the next section.

8.2. Digital campaign expenditures in Belgium: The numbers

As part of the research for this case study, data was collected on campaign spending for the different political parties that participated in the 2019 elections and had at least one seat in one of the federal or regional parliaments during the 2014–2019 legislature—that is, the Flemish socialists (sp.a, now Vooruit), the Flemish Christian Democrats (CD&V), the Flemish liberals (Open Vld), the Flemish nationalists (N-VA), the Flemish far right (Vlaams Belang), the Flemish Greens (Groen), the Walloon socialists (PS), the Walloon Christian Democrats (cdh, now les Engagées), the Walloon liberals (MR and Défi), the Walloon far right (Parti Populaire), the Walloon greens (Ecolo) and the unitary far-left party (PVDA-PTB). Table 8.1 summarizes the digital campaign spending of these parties both in absolute numbers (in euros) and as a percentage of their total campaign budget.

The Belgian party system is organized regionally, with the Flemish parties competing in Flemish-speaking constituencies and the Walloon parties in French-speaking constituencies. However, both Flemish and francophone parties compete in the bilingual constituency of Brussels. The far-left party (PVDA-PTB) is the only unitary party competing in both French and Flemish constituencies. Table 8.1 is therefore divided into different rows for Flemish, Walloon and unitary parties.

An analysis of campaign spending on the part of Belgian parties makes it clear that investments in more traditional tools still dominate. Only a little over a quarter of the total campaign expenditure of all parties together was spent on online ads. When looking at the Flemish and Walloon parties separately, the numbers change to 31 per cent and 23 per cent respectively. There are, however, three exceptions to this general trend. The Flemish far-right party (Vlaams Belang) is the clear leader, having spent almost 61 per cent of its total campaign budget on online ads on social media, leaving the other parties far behind. The Walloon liberals (MR) follow relatively closely, having spent 52 per cent of their total campaign expenditure on online ads, while the Flemish socialist party (sp.a., now Vooruit) also stands out, at 44 per cent. Additionally, Table 8.1 makes clear that only a marginal part of each party’s budget was spent on the creation of an election-specific website or webpage.

Turning to the candidate level (Table 8.2), the dominance of traditional campaigning is even more striking. Here, only 16 per cent of all 2019 campaign expenditures went to (targeted) online ads. Furthermore, this innovative tool seems much more popular among Flemish candidates, who spent 20 per cent of their overall budget on this type of advertising, compared with just below 6 per cent for their Walloon counterparts. Similar to the Flemish far-right party at the party level (Table 8.1), the Flemish far-right candidates stand out at the candidate level. Together, they spent 47 per cent of their campaign budgets on ads on social media, significantly more than the other parties’ candidates. The Flemish nationalist candidates find themselves in second place, having spent 19 per cent of their financial resources on online ads. Third place goes to the Flemish Christian Democrats (15 per cent), with the Flemish liberals (14 per cent), the Walloon far right (13 per cent) and the Flemish Greens (10 per cent) following suit. With only 9 per cent of their total budget dedicated to online ads, the tool does not seem very popular among the Flemish socialist candidates. This stands in stark contrast to the popularity of the instrument among the Flemish socialist party (Table 8.1), and the same goes for the Walloon liberal (MR) candidates. They spent only 4 per cent of their campaign budgets on online ads, compared with 52 per cent at the party level. The candidates from the other Walloon parties also spent a very small part of their budget on online ads. Candidates from the far-left unitary party (PVDA-PTB) do not have a tradition of individual campaigning, which is reflected in their reported zero spending. Table 8.2 also shows that individual candidates spent very little on an election-specific personal website or webpage.

Although the results of the 2019 concurrent elections fall outside the scope of this case study, the massive electoral gain of the Flemish far-right party should be mentioned. Compared with the 2014 concurrent elections, they obtained 15 extra seats at the federal level (the federal Chamber of Representatives consists of 150 seats), 17 at the Flemish level (the Flemish Parliament consists of 124 seats) and 2 at the European level (Belgium has 21 seats in the European Parliament). As soon as it became known that they had spent a large part of their campaign budget on social media ads—compared with the other parties competing in the 2019 elections—the public became convinced that their success was entirely due to the use of these (targeted) online ads. However, there is currently no scientific proof to support this idea because the research that specifically looks into the effect of online spending during the 2019 concurrent elections in Belgium is ongoing and the results are expected to be published by mid-2023. Furthermore, the hypothesis neglects the fact that the Flemish far-right party also spent almost half of its campaign budget on more traditional campaign tools. Nevertheless, the success of the far-right party and the ensuing popular belief are part of the context in which the legislative initiatives that will be discussed in the next section came into being.

8.3. Regulating digital campaigning and digital campaign financing initiatives in the Belgian context

The societal debate on the use of political ads on the Internet in general and on social media specifically has not escaped the attention of the Belgian federal and regional parliaments. However, most of the initiatives in this context have been proposed by members of the Federal Parliament. This is not surprising, as the federal level is responsible for the regulation of electoral campaigns for the European, federal and regional parliaments (see Section 1 for details). Local elections are covered by mandates at the regional level.

The first federal legislative initiative dates back to 6 October 2020 and was launched by six members of parliament (MPs) from the Walloon Socialist Party. They argued that there was no justification for banning political ads in cinemas or on television and radio during the official campaign period while ads on the Internet and in the press (i.e. newspapers and magazines) were permitted. Therefore, they proposed a ban on all commercial ads—both offline and online—during the official campaign period, which lasts four months. The initiators specified that flyers and leaflets handed out or distributed by the postal service do not fall under the proposed restrictions (Chambre des Représentants de Belgique 2020). It should be noted that the aforementioned 2013 decision to lift the ban on online political ads was taken based on a similar argument as the one used here to propose a total ban on both online and offline advertising (see Section 1). The proposed total ban makes this legislative proposal much more radical than its predecessor (Chambre des Représentants de Belgique 2007b), which banned online ads only. This initiative is currently part of a larger debate in the committee on institutional reforms in the Federal Parliament.

A second initiative that cannot be neglected is a resolution adopted by the Federal Parliament on 6 May 2021 that stresses the need for more transparency and accountability on the part of social media and other platforms dealing with online content and information. Regarding the specific issue of online political ads, the members of the Federal Parliament point to the vulnerability of the younger target audience especially with respect to the technique of microtargeting. They urge the federal government to plead at the European Union level for a code of conduct for sponsored political ads on popular social media platforms. This code of conduct should ensure that the identity of the actor behind a pushed and/or sponsored message is made public to the individuals receiving these posts (Chambre des Représentants de Belgique 2021a). This initiative seems to indicate a desire to transfer the responsibility for regulating online political ads to the supranational EU level.

The third initiative turns away from the official campaign period by specifically proposing an annual spending cap of EUR 1 million per political party (including their MPs and party personnel) on all political information dissemination activities and political ads, both online and offline. The spending cap mentioned in this legislative proposal would not apply during the official campaign period since Belgian campaign spending regulation already specifies and imposes a spending limit for both parties and candidates during that period. The initiators of the proposal, MPs from the Flemish and Walloon greens, argue that the proposed spending cap is necessary for three reasons. Firstly, it would help create a more level playing field between those parties that have a lot of financial resources and those that do not. The initiators stress that this objective was also behind the introduction of a spending cap during the official campaign period. Secondly, the MPs in question argue that, if passed, the proposal would put an end to the increasing amount of financial resources spent on social media and other online and offline ads by (certain) political parties outside the official campaign period. Finally, the MPs pushing the proposal point to the relationship between microtargeting and the development of online echo chambers; the latter limit the exchange of (conflicting) ideas, which is of paramount importance to the democratic process (Chambre des Représentants de Belgique 2021b). Like the first initiative discussed in this section, this legislative proposal is currently also part of the larger debate in the committee on institutional reforms in the Federal Parliament.

The last initiative is taking place at the regional level, in Wallonia. As already mentioned in Section 1, local elections (i.e. in municipalities and provinces) fall under the mandate of the regions in Belgium. Consequently, local campaign regulations differ from region to region. Electoral campaigns at the Walloon level face similar restrictions as campaigns for European federal and regional elections. During the official campaign period, which starts three months prior to election day, ads are prohibited on both radio and television, and the same goes for the distribution of small gifts, the use of commercial billboards and commercial campaigns over the phone (Parlement de la Wallonie 2004: article  L4130-4). In May 2022 the Walloon Government agreed on a draft law to amend the Democracy Code. Among other changes, the government proposes limiting campaign spending committed to targeted messages on the Internet to 50 per cent of the total amount of money a candidate or a party is allowed to spend according to their respective spending limit. The Walloon Government argues that this restriction is necessary to protect the personal data of citizens, as paid messages on social networks are often aimed at collecting recipients’ personal information, according to the initiators of the amendment. Furthermore, the Walloon Government argues that the proposed measure is necessary to limit the spread of fake news, as this phenomenon is deemed to threaten the liberty of the population as well as democratic institutions.⁶ This draft law will be sent to the Council of State and then to the Walloon Parliament, where it will be discussed and possibly adopted.

8.4. Conclusion

The societal debate on regulating political messages on social media platforms is ongoing at all levels of Belgian politics. However, the initiatives discussed in this case study suggest that Belgian politicians are turning to the EU for new legislative initiatives on (targeted) political messages on the Internet. After all, the resolution of the federal Chamber of Representatives of 6 May 2021 is the only initiative which has been adopted by the Federal Parliament so far. It is true that the other federal initiatives are still being discussed in the relevant committee; however, the fact that their initiators belong to different parties of the governmental coalition seems to indicate a lack of agreement on the issue within the majority. The probability that these federal legislative proposals will be adopted is therefore rather small. The contrary is true for the initiative by the Walloon Government, as all parties in the Walloon governmental coalition seem to support the draft law. However, the impact of this legislation will be limited to local electoral campaigns in the Walloon region.

Key takeaways:

• In Belgium, spending on traditional campaign tools exceeds spending on online tools in most cases. However, some political parties have recently spent large amounts of financial resources on their online communications, which may incentivize others to follow suit.
• In Belgium, the main approaches (some are still under discussion) to regulating party spending on digital campaign tools are (a) banning all political ads during election campaigns; (b) placing spending caps on online advertisements; and (c) advocating for an EU-level code of conduct for sponsored ads on social media platforms.
• At the same time, there are diverging political views over these approaches, and some may not materialize.

Amaro Grassi

9.1. Introduction

In Brazil, campaign finance is an ongoing issue that comes into focus with each electoral cycle. In recent history, there have been various different arrangements put into place to fund the electoral process, and the changes made have often reflected the interests of the political elite at the time. Although, in many respects, the Brazilian electoral system has changed little since the first republic (1889–1930),⁷ the way in which electoral finance works has never followed exactly the same rules, and each election has seen adjustments in how candidates and parties can raise and spend money.

In 2017, for the first time, organic electoral advertising on social media,⁸ social media boosting and paid promotion in web search engines were included on the list of permitted expenditures for electoral campaigns. The Internet, however, had long been used as a tool to acquire votes, although without the appropriate regulation. Including digital social media as an option for electoral campaigning by candidates and parties has been beneficial to the electoral system—both in terms of competition, as political actors now have another legal tool through which to reach potential voters, and in terms of transparency, because candidates are now required to account for this type of campaign spending. Nonetheless, there are still matters that require consideration and evaluation—for example, the possibility of an imbalance in the campaign arising from economic disparities in the competition—as the legislation lacks some features, such as a spending limit.

Considering the above, the goal of this case study is to present information on Brazil’s current framework for regulating the nexus between digitalization and campaign finance. It focuses on the use of digital tools for political campaigns, the lessons learned and the outstanding regulatory gaps. The discussion concludes with some proposals for electoral stakeholders to take into consideration.

9.2. Political and legal framework

Brazil has a majority electoral system for executive positions in the municipal, state and federal spheres, as well as for the Federal Senate. In turn, the country operates an open list proportional model for the legislative positions of city councillor, state deputy, district deputy (for the Federal District, Brasília) and federal deputy in the Chamber of Deputies. This means that parties do not put candidates on the list in a particular order; instead, the election results are defined by the voting, with candidates ranked according to the number of votes received from the electors in descending order.

Democracy in Brazil has developed under this framework since the Constitution of 1988. However, this development has not followed a linear path, and there have been various attempts to implement a different system, including a parliamentary system, district voting, verticalization,⁹ and the possibility of private (corporate) financing of campaigns, to name a few. Regarding money in politics, two dimensions are worth emphasizing: (a) how money gets into politics; and (b) how money can be used in campaigns. In Brazil, both are regulated by electoral legislation and norms developed by the Superior Electoral Court (Tribunal Superior Eleitoral, TSE). These regulate who can donate to parties and candidates, which then affects the strategies available to the political actors in raising campaign funds, and how much money can be spent on elections, which determines the strategies that can be used to acquire votes.

Having previously been banned, corporate financing was allowed, by Law 8.713/1993, after the PC Farias scandal, which led to the impeachment of the then president, Fernando Collor, in 1992. The law stipulated clearer rules for private sector action in electoral campaigns, including donation limits by both individual donors and legal entities. Corporate financing was banned again after the Mensalão scandal in 2005 and the Lava-Jato scandal in 2014. From these scandals, the notion emerged in the public mind that corporate finance, in addition to undermining the ‘one man, one vote’ principle of liberal democracy by introducing inequality in democratic participation, was also a significant source of corruption in the country (Campos and Peixoto 2015). While corporate financing of political parties has remained banned since 2015, donations from citizens (individuals) and self-financing by the candidates have both been maintained. A 2018 political reform created the Special Campaign Financing Fund (Fundo Especial Financiamento de Campanha, FEFC), a major provider of public funds to political parties, distributed based on their parliamentary representation. The parties also have access to the Party Fund, a smaller source of public funding created in 1965 (available to parties on an annual basis, including in election years).

9.3. Digital campaigning in Brazil

In every election in Brazil since the year 2000, the use of the Internet in electoral campaigning has intensified but, undoubtedly, this has seen an exponential growth in recent years, alongside the rise in the general use of social media. The TSE regulates the elections using norms, such as resolutions,¹⁰ in the absence of decisions by the legislative branch. Table 9.1 provides a summary of the main legal changes in each electoral cycle. In terms of spending on Internet promotion, this was standardized by the TSE in the absence of a legislative process until 2009 and Law 12.034/2009 (Brazil 2009). The Law prohibited paid promotions on social networks and search engines. However, in 2017, paid promotions for electoral campaigning on Internet sites and social media were introduced into the legal system through Law  13.488/2017 (Brazil 2017). Since then, three elections have taken place (2018, 2020 and 2023) wherein political actors have been permitted to mobilize digital media for electoral campaigns.

Brazilian legislation does not set any specific spending limits for online campaigns, aside from the general campaign spending limits (these vary according to the elected office being contested; for a complete table by position in each federation unit, see TSE 2022). In general, there are no limits on the types of expenditure, except for meals for staff providing services to candidates, and automobile rentals, which are limited to 10 and 20 per cent of total campaign expenditure, respectively. Table 9.2 shows the allowed funding sources, and Table 9.3 the permitted and prohibited types of expenditure.

Parties and candidates must notify the electoral management body, using an electronic reporting system, of the amount spent, provide a description of the expenses, information on the service provider and a copy of the invoice (receipt). Data on electoral spending is drawn from these submitted accounts.

9.3.1. 2022 elections

Brazil held elections in 2022, when more than 156 million voters were eligible to vote for the positions of president and vice president, in addition to the positions for governor, vice governor, senator, federal deputy, state deputy, and district deputy, contested by over 28,000 candidates. Approximately 20 per cent of the eligible voters abstained from voting. There were 11 candidates in the presidential race, with 2 ending up in a run-off—Luiz Inácio Lula da Silva of the Partido dos Trabalhadores (PT) and Jair Bolsonaro of the Partido Liberal (PL). The former secured a narrow majority of votes and was elected president, making him the first Brazilian president to serve a third term. In addition to the presidential race, 32 parties put forward candidates for other positions. The high number of parties resulted in a Chamber of Deputies consisting of 23 elected parties; that is, elected representatives from across 23 parties occupied the 513 legislative seats as federal deputies.

Of the total funds for electoral campaigns in 2022, around 81 per cent were from public sources (FEFC 73.9 per cent and Party Fund 7.5 per cent). Donations from individuals (citizens) accounted for 12.43 per cent, with 0.32 per cent from crowdfunding. Self-financing accounted for 2.94 per cent, and donations between candidates accounted for 2.96 per cent.

In 2022, the spending limit for the office of president was approximately EUR 16 million for candidates in the first round, plus EUR 8 million for candidates reaching the second round.

Two rules were in force in the 2022 elections regarding prohibited periods for campaigning: (a) two days before the election until election day (30 September to 2 October),¹¹ with the prohibition of paid promotions, including on the Internet; and (b) one day before the election until election day (1 to 2 October), with the prohibition of printed materials. The prohibited period for electoral campaigning is ambiguous, since it is not clear whether organic (not paid) advertisements are still permitted, including on social media platforms.

According to the PubliElectoral study carried out by Fundação Getulio Vargas—Escola de Comunicação, Mídia e Informação (FGV-ECMI) and Asociación por los Derechos Civiles (ADC), 171 of the 375 accounts monitored did not comply with the period for paid electoral advertisement in the first round (ADC and FGV-ECMI 2023). Regarding organic advertisement, 306 of the 375 accounts did not comply with the prohibited period in the first round, possibly because the rules were not clear regarding its application to virtual environments. This regulatory gap regarding the permitted period of advertisements social media could potentially give competitive advantage to those parties who do not comply. In addition to clarifying the permitted period of advertising on social media, including paid and organic, it is equally important to carry out effective monitoring, which could be facilitated by the implementation of ID and user information for candidates.

Although the electoral reporting system has a considerable degree of transparency, it is important to note that around 20 per cent of the first-round candidates did not declare their expenses. But because they were unsuccessful, exemptions in Brazil’s current system for non-elected candidates relieve them of retrospective reporting obligations.

The main change brought by the TSE’s Resolution 23.714/2022, implemented in the middle of the electoral process (20 October 2022), sought to reduce the potential impact of fake news on election results by expediting the process of removing it from circulation. Previously, a lawsuit had to be filed for each link suspected of containing fake news, even if different links had the same content. The resolution enabled legal decisions to be used to target any existing or future link containing the same content.

9.3.2. Trends in political actors’ spending on social media

Based on available TSE data for the 2018 and 2022 general elections and the 2020 municipal elections, the total campaign spending amounted to EUR 730 million, EUR 1 billion and around EUR 695 million, respectively. As shown in Table 9.4, an increasing amount was spent on social media advertisement with each election: EUR 19 million in 2018 (2.6 per cent of declared spending); EUR 24 million in 2020 (3.47 per cent); and EUR 69 million (6.0 per cent) in 2022.

Spending on social media has grown with each election at the same time as the proportion spent on radio, television (audiovisual production) and sound trucks/cars has fallen, with the former almost catching up with the latter in 2022. The biggest expense in the three elections is printed advertising (stickers, newspapers, magazines and various printed matter). Events (rallies, street mobilization, candidacy events and others) account for a significant part of candidates’ expenses but dipped in 2020 during the Covid-19 pandemic, when the electoral management body introduced some restrictions on in-person events. In total, direct campaign advertising spending amounts to more than 50 per cent of campaign spending in Brazil.

Table 9.5 shows the online paid promotion expenditure broken down by digital platform over the three elections analysed. Taking the three elections into account, the platforms Meta (Facebook and Instagram), Google (Google and YouTube), TikTok and Kwai were used for paid promotions.¹² Meta was the largest platform for political campaigning, accounting for more than 80 per cent of the expenditure on paid promotions in the 2018 and 2020 elections, dropping to 60 per cent in 2022. In 2022, Google reached 33 per cent of the expenditure on paid promotions, while TikTok and Kwai gained some presence, but did not reach 1 per cent.

It is important to note that social networks also include instant messaging apps, such as WhatsApp and Telegram. Although these tools do not permit paid advertising, they are part of electoral campaign strategies. As demonstrated by Piaia and Alves (2020), the flow of messages in WhatsApp groups supporting Jair Bolsonaro is by no means spontaneous. Rather, they reflect the presence of ‘super participants’—individuals accounting for a high percentage of posts and interactions, executing a coordinated strategy. A report published in the newspaper Folha de São Paulo shows evidence of millions spent by entrepreneurs in mass messaging services (Campos Mello 2018). If confirmed, this amounts to an illegal strategy since, as mentioned, current regulation prohibits companies from donating to political campaigns.

9.3.3. Role of influencers and digital marketing firms

A frequent strategy in Brazil is the recruitment of influencers (performing artists, television presenters, radio commentators) not only to endorse candidates but also for them to stand as candidates. Called ‘vote pullers’, they have the capacity not only to get themselves elected but also to help other candidates in the same party get elected too. In Brazil, all votes for a candidate count towards the total for their party. Thus, a popular candidate can obtain enough votes to get her- or himself elected as well as others from the party/coalition/federation. For example, suppose that a party needs 1,000 votes to enable one delegate to be elected. If the party has five candidates, and one individual candidate receives 5,000 votes, then all five candidates can be elected.¹³ In this case, four candidates were elected by ‘vote pullers’ (Carlomagno 2016). To remedy the situation, Brazilian legislators adopted a barrier clause to reduce this vote-pulling phenomenon.

Influencers continue to exercise influence over audience behaviours (in this case, voting behaviour), through their accessible, emotive and/or humorous public image. For instance, a survey carried out in Argentina, Brazil, Colombia and Mexico in 2022 identified that young people aged 16 to 24 are informed about politics through supposedly non-politicized channels and profiles, as they get distanced from traditional politics and its partisan intermediaries (Luminate 2022).

While presenters on traditional broadcast media (such as television or radio) are not allowed to promote electoral candidates, the legislation (Law 13.488/2017; see Brazil 2017) allows digital influencers to campaign for their favourite candidate, as long as they are not paid to do so. Paid electoral campaigns on the Internet are exclusive to the candidates and parties/coalitions/federations involved in the election. Additionally, the TSE Resolution 23.671/2021 prohibits the hiring of individuals or legal entities that circulate paid advertising of a political-electoral nature in the profiles, pages, channels—or more broadly in digital environments—of the political candidates in electoral competitions.

Although there is no evidence of influencers being hired to do electoral campaigning, the 2018 and 2022 elections saw influencers having a massive presence in the campaign generally. For instance, the 2018 election was marked by the presence of digital influencers—including some who were elected, such as Joice Hasselmann and Alexandre Frota (Partido Social Liberal São Paulo (PSL-SP)), both new to representative party politics, who emerged amid the demonstrations in June 2013 and the impeachment process of President Dilma Rousseff with an anti-establishment position. This became even clearer in 2022, when parties sought to recruit digital influencers for candidacies and campaigns, such as the candidates for Federal Deputy, Nikolas Ferreira (PL-Partido Liberal Minas Gerais) and André Janones (Avante-Partido Liberal Minas Gerais), both of whom were architects of digital campaigns for Bolsonaro and Lula respectively and were at some point subjected to platform sanctions after flouting the rules for Instagram and Tiktok.

The electoral legislation also establishes transparency requirements—for instance, each electoral advertisement must be identified with the candidates’ corporate taxpayer number (Cadastro Nacional de Pessoa Jurídica, CNPJ) and information on the party/coalition/federation. However, the transparency tools for platforms such as Meta and Google do not follow a common standard, which makes it difficult for the electoral management body, citizens, stakeholders, research institutions and civil society to monitor the information.

In addition to this lack of standardization, the monitoring of the 2022 elections carried out by FGV-ECMI and ADC, within the PubliElectoral project, found that the official reports provided to the TSE do not allow the identification of which pages and accounts were boosted; it was only possible to identify which service happened on which platform (ADC and FGV-ECMI 2023). In this context, any legislation improving transparency is welcome, particularly in an electoral system with large amounts of public money in circulation.

9.4. Conclusions and lessons learned

Electoral transparency is key for the proper functioning of democracy, especially where systems are unequal. Since 2016, Brazilian electoral campaigns cannot receive funds from legal persons. However, the system is still deeply unequal, due not only to the distribution system of the FEFC but also to the Party Fund: both funds benefit parties with a higher number of representatives from the previous election, which favours those that are already established.

Electoral campaign finance in Brazil enjoys a certain level of transparency overall. However, there is a need to improve the specific regulations for campaigns in digital environments. These improvements could provide more transparency, increasing the legitimacy of the Brazilian democratic process. Considering that most of the resources spent on campaigns are public (from the FEFC and Party Fund), any regulation that makes electoral expenses more transparent is also a step forward in the accountable use of public funds.

The allowing of paid promotions in digital electoral campaigns in Brazil was important for candidates and parties, to enable them to have another tool as they seek votes. However, it was also important for the system’s transparency, increasing the level of traceability compared with alternatives that tend to remain under the radar, such as artificial boosting, when bot strategies are used to create a higher volume of interaction and to ‘force’ the platform algorithm to deliver content to more users.

There are currently at least 42 bills pending with the legislative branch, including Bill No. 2630/2020, known as the ‘fake news bill’, and Bill No. 9.532/2018, which is also related to norms on potential fake news. The former has advanced further and almost went into a vote in the middle of May 2023, but was removed from the agenda due to negative reactions in several sectors of society. These bills have a direct impact on the electoral process, in terms of both regulating the behaviour of big tech and providing more transparency—for example, the ability to identify who has contracted a specific social media boosting campaign by corporate taxpayer number (CNPJ) or individual taxpayer number (Cadastro de Pessoa Física, CPF).

Bills such as Bill No. 2630/2020 have mobilized some sections of civil society, the private sector and specialists, who have pressured public authorities to ensure that digital environments are democratic spaces. In this way, disputes around the guidelines stated in the legislation gain visibility in the public debate, such as the fight against disinformation and hate speech, the defence of freedom of expression, and policies for the remuneration of journalistic content, as well as transparency in the policies that govern the use of social media networks.

Considerations for the future

In light of the above, electoral stakeholders in Brazil are encouraged to consider the following:

1. Ensure there is traceable identification of paid online promotion—by individual (CPF) or corporate (CNPJ) taxpayer number. It is important to note that more transparency is needed in terms of platforms being able to identify who spent money on social media and online advertisements in elections, through traceable CPF and CNPJ taxpayer numbers.
2. Level the playing field (in terms of prohibited time periods) between organic digital campaigns and traditional, offline campaigns. It is important to create laws that clarify the limits and possibilities for organic digital campaigns, as well as paid promotions. If candidates must comply with a prohibited period for campaigning in the physical world, so that those with more resources are not favoured in the days immediately before the election, the same system must apply to the virtual world, so that those with more funds and/or engagement on each platform are not favoured to the detriment of others.
3. Enforce accountability for campaign finance, regardless of whether the candidate wins or loses the electoral contest. Losing candidates are not fully accountable for their electoral finance, hence it should be political parties’ prerogative to require candidates to submit their accounts. Since every winning seat in the legislature belongs to the party, they must face sanctions if their candidates are not adequately accountable.
4. Work with digital platforms to block coordinated mass-messaging by ‘individual’ accounts. As corporate spending is prohibited in the Brazilian electoral system, when it does take place, it is happening under the radar. Therefore, one of the ways to deal with the problem of corporate spending on political advertising is to work in partnership with the digital platforms, to reduce the number of messages that each ‘individual’ can send during an election period and also to identify the source and destination of massive electoral publicity, so as to block it and implement sanctions, if necessary.
5. Encourage civil society involvement in the monitoring efforts of the electoral management body, the TSE. While the TSE has in-house data processing capacity to monitor spending online, its efforts can be complemented with participation of civil society in active monitoring of online spending by political actors.

Maria Jaraquemada

10.1. Introduction

Chile has experienced five electoral campaigns in recent years—at least one per year from 2020 to 2023. Since half of these took place under pandemic conditions, digital campaigning featured heavily, as a result of social distancing measures. In the 2021 presidential elections, digital advertising—social networks, automated calls, mass text messages and the use of artificial intelligence—replaced traditional methods of campaigning, such as door-to-door canvassing, rallies, and billboard campaigns, to a large extent (Olave 2021).

Changes were made to the regulation of electoral campaigning and advertising in Chile in 2016, in response to scandals involving political finance. These were the most significant reforms since political finance regulations were first adopted (in 1988, during the dictatorship and subsequently modified in 2003). However, the 2016 reform failed to keep pace with the explosive growth in new technologies and campaign tactics.

10.2. Political and legal framework

Chile has a presidential system, where the president—who is both head of state and head of the government—is elected for a four-year term and cannot be immediately re-elected. The parliament is bicameral, comprising a Chamber of Deputies with 155 representatives and a Senate with 50 members, both chosen by a party list proportional representation system and D’Hondt coefficient (IPU n.d.). Currently, 21 different political parties make up the Chamber of Deputies, a number that has increased in recent years, especially since changes were made to the electoral system in 2016. Between 2012 and early 2023, voting was voluntary; compulsory voting was reintroduced through a 2022 reform, applicable for the first time during the election of the constitutional councillors held in May 2023. Out of the total population of approximately 19 million, more than 15 million are registered to vote.

Chile’s first regulation on political campaigning, Law No. 18700 (Chile 1988), was introduced by the military junta. It contained a few rules on electoral campaigning: funding could only be sourced from within the country; campaigning could be carried out via the press, radio stations and television channels, as well as by means of flyers, banners or notices; and campaigning was permissible from the 30th, and up to the third day prior to an election or plebiscite (Chile 1988: para. 6). In 2003, after scandals involving the misuse of public funds for financing campaigns, known as the MOP-GATE case (see, for example, TV Poder Judicial 2018), a new regulation was adopted: Law No. 19884 (Chile 2003). This new law outlined stricter regulations on the origin of funds, established a mixed system with private and public funding, and placed limits on donations and expenses. It also required more transparency in private funding and instituted administrative sanctions—mostly fines—for any infringements.

Some years later, with fresh cases related to illicit political financing, known as Penta and SQM (see, for example, Laborde 2023), and evidence of regulatory weaknesses, some modifications were adopted by means of Law No. 20900 (Chile 2016). This followed the recommendations of an Advisory Committee to President Michelle Bachelet, known as the Engel Commission (after its President, Eduardo Engel) (Engel Commission 2015).

Related to campaign finance, the main substance of the current regulations is as follows:

1. Regulated periods for campaigning. Most campaigning starts 60 days before elections, but propaganda in public and private spaces with posters and banners can only be displayed 30 days before (in authorized spaces and with size restrictions). Only for presidential elections can ‘pre-campaigning’ take place 200 days before elections. During this period, candidates can receive funding and incur expenditures for campaigning.
2. Registry of electoral supporters. Referred to as ‘brigadists’, these are personnel who either volunteer or are hired to support political campaigns. Candidates are liable for any damages caused by these individuals, and people with a criminal record are barred from becoming brigadists.
3. Mixed system of funding. The state provides an advance to political parties according to previous electoral results, as well as a subsidy for free television spots and reimbursement of expenses according to the number of votes obtained in the current election. Companies and legal entities, such as non-governmental organizations (NGOs), can no longer contribute private campaign and party funding—it must come from individuals only.
4. All contributions are made public. Details of all contributions, other than minor ones,¹⁴ are published regularly on the website of Chile’s Electoral Service (Servicio Electoral de Chile, SERVEL).
5. Definition of electoral expenditure. This is understood as any disbursement or contribution in cash, made by the pre-candidate, the candidate, a political party or a third party in their favour, on the occasion and for the purpose of electoral acts.
6. Limits on electoral expenses¹⁵ and on individuals’ contributions¹⁶. All contributions and expenses must be made to and from a bank account for this sole purpose, controlled and monitored by the Electoral Service.
7. Role of the electoral management body. The Electoral Service, an autonomous entity, is responsible for oversight. There is a specific form—Form 105—to report expenses relating to digital campaigns. Reporting channels for possible legal infringements are in place.
8. Advertising in violation of the law is punishable with fines. Fines are up to five times the excess over election spending limits, but in some cases, serious offences can be sanctioned with loss of public office; imprisonment of up to five years; or lesser, non-custodial sentences.

The Electoral Service publishes guides for each election, providing details on donations and expenses. The first such guide after the 2016 legislation reflected a very restrictive view by stating that digital campaigning was forbidden (see, for example, El Mostrador 2016). This came as a surprise, both to the public and to political parties; some Deputies even proposed a bill to address this issue. As a result, the Electoral Service had to retract and clarify their criteria (Zamora 2020).

10.3. Patterns in digital campaigning

Over the last few years, there has been an increasing mobilization of online channels by political contestants to reach their constituencies. Lionel Brossi, from the Faculty of Communication and Image at the University of Chile, and a member of the Artificial Intelligence and Society Hub, commented that the importance of social networks is increasing with time: ‘They enable the possibility of sending messages to different segments of the electorate … [microtargeting] is a marketing strategy that uses data, very detailed categories of individuals or the population in terms of their interests, their preferences, to influence their decisions’ (Tapia 2021).

10.3.1. Presidential elections in 2017 and the 2020 referendum

The civil society organization Digital Rights (Derechos Digitales) has been researching the impact and use of digital media advertising in political campaigns in Chile since 2020. In its first report (ADC and Derechos Digitales 2021), it analyses the use of digital media in the 2017 presidential elections, which suggests that the percentage of electoral expenses allocated to digital and social media was still very low (see Figure 10.1). Candidates spent, on average, 4.2 per cent of their total electoral expenses on digital media. It must be noted that this information comes from the reports provided by the candidates, which are approved by the Electoral Service.

In the initial referendum on the constitutional process in 2020, the same study shows that, on average, the platforms of political parties or the candidates published 97.7 per cent of their electoral advertising during electoral periods (ADC and Derechos Digitales 2021: 35). By the same token, social networks are mainly used for electoral campaigning (propaganda) during these periods, rather than for other purposes. This might seem obvious, but the pattern as reported has not been the same for all recent electoral events.

10.3.2. Presidential elections in 2021

Derechos Digitales’ analysis of the 2021 presidential election shows that all candidates used Meta (the parent company of Facebook, WhatsApp and Instagram) for political advertising. During the first round, with six candidates competing, most of the content posted on the monitored accounts (62 per cent) were categorized as ‘non-electoral’ (not fully related to the campaign); only 36 per cent (2,238 publications) were ‘electoral’ publicity; and the remaining 2 per cent were categorized as ‘doubtful’ (Figure 10.2) (ADC and Derechos Digitales 2022). For the purposes of the study, the advertising analysed included both the registered candidates’ official campaign outputs, and those of ‘satellite accounts’ belonging to entities with public influence—institutions, party factions and individual figures—who had publicly declared their support for any of the registered candidates. At least during the first round, which took place at the same time as the parliamentary election, the percentage of ‘electoral’ publications was inversely proportional to the size of the party (ADC and Derechos Digitales 2022: 35).

In the second round, the numbers changed dramatically, as 59 per cent of advertising on Meta was categorized as electoral and 37 per cent as non-electoral (ADC and Derechos Digitales 2022: 39). The study concludes that this platform is a significant space for electoral campaigning and that, as with offline media, the rules about the permitted periods for electoral campaigning are not necessarily respected on social media (ADC and Derechos Digitales 2022: 44). Claims about campaigning outside permitted timelines are a regular feature of elections in Chile (see, for example, Radio Universidad de Chile 2020; Lingua 2022) and the Electoral Service faces oversight challenges specific to social media.

The Derechos Digitales study finds that, in 2021, not all political parties or presidential candidates reported their digital expenses using the stipulated Form 105. In some cases, the correct form was used to report digital expenses but it lacked the required detail about specific platforms and services purchased. Also, in almost all cases, the spending data given on the form did not match the data published by Meta, considering that political parties and candidates may have reported expenses made on other social media networks, but their ad libraries are not available (ADC and Derechos Digitales 2022: 54).

Unusually, one of the first-round candidates, Franco Parisi, was not physically in Chile in the pre-election period—due to unresolved legal issues related to pending alimony—so his campaign was mainly carried out by digital means, especially through a weekly YouTube programme called ‘Bad Boys’ (which, according to YouTube and Google (n.d.b), is not considered advertising content). Using such platforms, Parisi obtained third place in the first round of the presidential election, with 13 per cent of the vote. This case demonstrates that there is a challenge in accounting for expenses made by candidates campaigning from outside the country. National political finance regulations and oversight rules often only apply within the country and, consequently, it is difficult to control the misuse or underreporting of campaign expenditures incurred in another country (Figueroa 2023).

Figures 10.3 and 10.4 give a breakdown of officially reported expenses in the second round by type of electoral propaganda and by two presidential candidates, respectively.

10.4. Challenges related to online campaign finance in Chile

10.4.1. ‘Electoral propaganda’—definitional issues

To understand the situation regarding digital campaigns, it is very important to consider the current definition of ‘electoral propaganda’ in law (as modified in 2016):

[A]ny public event or demonstration; or radio, written, visual, audiovisual or other similar media advertising, if it promotes one or more persons or political parties, for electoral purposes. In the case of plebiscites, propaganda shall be understood to be that which induces support for any of the propositions submitted for consideration by the citizens. Such propaganda may only be carried out at the time and in the manner prescribed by law … dissemination of ideas or information on political acts carried out by natural persons shall not be understood as electoral propaganda. 
(Chile 2016: article 30; author’s translation)

Considering the above, the following observations can be made in relation to campaigning through social networks or social media in Chile.

When paid services are commissioned on social media (such as Facebook or TikTok) to promote a candidate, this is advertising and falls under the definition of electoral propaganda, and hence should be governed by the corresponding rules and declared as an expense. Where a person or agency is hired to plan or execute viral communication strategies, this is defined in the same way and the same rules apply. Also, if a person offers their services for free, these must be declared and valued as contributions.

In the case of telephone and email campaigning, the same criteria as above apply, regarding advertising and contracting communications services. Activities of this type should not be considered mere private communications, as they involve expenditure on publicity for electoral campaigns—and should be regulated accordingly.

However, the law requires (Chile 2016: article 16) that stakeholders distinguish between promotional activities (what might be called political marketing or messaging) and those merely ‘disseminating ideas or information on political acts or expressions’, with the latter treated as an exercise in freedom of expression. This is a grey area and where to draw the line is sometimes not at all obvious. Ultimately, it is up to the Electoral Service to determine where to draw the line, as it exercises its oversight powers, and on a case-by-case basis if and when it receives complaints. For example, if a social media influencer uses their personal accounts to promote a candidate, and no advertisement is purchased, it is usually considered part of their right to exercise freedom of expression. But this is an example of where the Electoral Service would need to determine whether the activity crossed the grey line, as it could be considered an in-kind donation of services, such as a performance by an artist during the closing act of the election campaign might be.

In the many guides published for each election since 2016, the Electoral Service has clearly stated that digital advertising expenditure (i.e., on digital or social media, including social networks) must be declared as part of electoral expenditure. This must include information on the platform used, the contract period and the amounts involved. These guides state that on social media platforms only ‘promotional advertisements’, involving the commission of paid advertising services, are considered electoral propaganda, and not the communications from the personal profiles of candidates or political parties (for example, SERVEL 2017, 2021c).

The regulation described above was considered insufficient by the Congress; therefore, for the extraordinary referendum on the 2021 constitutional process, a special rule was established in the Constitution, just for that process. In addition to the application of some general rules, there were provisions for social networks. These stated that, although political expressions made through personal or group social media profiles do not constitute electoral propaganda (and would instead be protected by the freedom of expression guarantee), commissioning political advertising services through digital platforms would be considered electoral propaganda, and therefore be subject to the general campaigning restrictions.

10.4.2. Social media platforms’ own policies

Following the Cambridge Analytica affair (Confessore 2018), Meta now publishes information related to political advertising on its platforms, which is available to all (Meta n.d.), as well as sharing these data with the Electoral Service directly.

Meta uses a broader definition of political advertising than the Chilean law on electoral propaganda. According to Meta, political advertising is anything that:

• Is made by, on behalf of or about a candidate for public office, a political figure, a political party, a political action committee or advocates for the outcome of an election to public office.
• Is about any election, referendum or ballot initiative, including ‘get out the vote’ or election information campaigns.
• Is about any social issue in any place where the ad is being run
  (Meta Transparency Center n.d.).

As a result, campaign activity referenced in Meta’s Ad Library is correspondingly inclusive.

TikTok, on the other hand, excludes anything it defines as political content from permitted—paid—advertising:

Political content is not allowed in any form of advertising … this includes, and has always included, both brand ads and paid branded content. … Sharing political beliefs and engaging in political conversation is allowed as organic content, but it cannot be used with paid ad products, such as Spark Ads and Promote.
(TikTok n.d.)

Google has different requirements for political and election advertising based on region. In some regions, election ads may run only if the advertiser is verified by Google. In regions or countries where election advertising is required to be verified, the advertisement must contain information on who funded it: an advertiser ID label ‘is generated from the data provided during the verification process and is automatically included in most ad formats’ (Google n.d.a).

Further restrictions may apply according to local regulations, but this appears not to be the case in Chile. All the information is published in Google’s Transparency Center (Google n.d.b).

In the case of Telegram, political adverts are forbidden: ‘Ads must not promote political campaigns, elections, political parties, candidates, political or religious movements’ (Telegram n.d.).

As seen above, different platforms treat political advertisements differently which has a practical implication on how expenses are reported and accounted for. This creates omissions and often underreporting of expenditures online, thwarting the goals of attaining transparency in campaign finance.

10.4.3. Challenges in monitoring and compliance

One of the difficulties in the oversight of digital campaign advertising in Chile is that it mostly relies on reporting from candidates and political parties (Castañón 2023). Most provide information about their recruitment of communications and public relations agencies, but without disclosing which social media platforms are used and what specific strategies are employed—for example, ‘bots’, ‘troll farms’, indirect campaigns, etc. (Castañón 2023). Through its Communication Unit, the Electoral Service monitors activity on social networks, using specialized professionals and technology. It is aware of the difficulties of this task and the need for more capacity, especially given the trend towards greater levels of digital campaigning (Figueroa 2023). A related challenge is that there seems to be no accountability obligation from political actors for political advertisements placed by these outside agencies. On the one hand, political parties and candidates have the incentive to outsource digital advertisement services; on the other, they refuse to take responsibility for information or statements disseminated by these agencies even if there are clear similarities in the messages conveyed (Castañón 2023; Figueroa 2023).

A second difficulty—not unique to Chile—is, as highlighted previously, that Internet and social media companies self-regulate themselves, often in distinct ways, especially in the absence of national legislation. Additionally, they may operate in a national territory without being legally established there so there is a lack of jurisdiction over them. This means that there is no real control or oversight by electoral bodies in the region, even less when there are attempts to decrease the budget and personnel of these oversight institutions or to diminish their autonomy (as happened in Mexico and Brazil, for example).

Third, it is very difficult—if not impossible—to track digital advertisements in detail and to trace who paid for them (‘following the money’) and therefore to determine if spending limits have been complied with. Some of the expense categories identified, in reviewing the election expenditure reports for the presidential second round (SERVEL n.d.), included ‘Services from costume designer for the second round of the presidential campaign’ and ‘Protagonist and extras from the second round of the presidential campaign’, and others as general as ‘Publicity for the 2021 presidential election campaign second round’ or ‘Social media management support second round presidential campaign’. Such vague expense categories make it extremely difficult to determine whether contribution or spending limits were complied with, or whether any illegal means of campaigning were employed.

10.4.4. Indirect campaigning and constitutional processes

The loopholes in Chilean regulation can also be considered in relation to indirect campaigning: advertising or propaganda coming from campaigning movements, grassroots or social organizations for, or against, causes that are related to a specific candidate or party’s manifesto. Campaigns carried out by these bodies, as per the definition of electoral propaganda by the Chilean Electoral Services, may be considered manifestations of freedom of expression, as happened with super political action committees (PACs) in the United States (Zamora 2020). CIPER Chile, an investigative journalism NGO, showed how these loopholes were used in Chile without any oversight (Segovia, Sepúlveda and Herrera 2022). In some cases, social movements or grassroots organizations effectively promoted electoral propaganda through these platforms without prior authorization or registration, resulting in no accountability or transparency. In other cases, digital platforms considered some projects promoting access to information or citizen engagement with the constitutional process to be political advertising, even though they were following national regulations and not promoting specific candidates or parties, or any alternatives to the referendum. This points out the need to adopt some criteria that considers the particularity of social media and that allows for better oversight.

The two referendums carried out in Chile in 2022 and 2023 showed how—for this kind of electoral process, even more so than in conventional electoral competition between candidates—it was difficult to track who was behind various campaign initiatives. Some of these involved promoters of the different constitutional options paying for advertising on social networks without having been registered with the Electoral Service to carry out electoral propaganda. Some research has suggested that such unregistered (i.e. unauthorized) expenditure on social networks outweighed official campaign spending (Garay 2022).

When reviewing the list of political, social or electoral advertisers on Facebook and Instagram, it is possible to see that many of the pages that have invested the largest sums of money in referendum campaigning are not registered to do so. Not all social media networks share such information with the public bodies in charge of monitoring this, as Meta does. Even where companies do publish this information (such as Meta), there are also difficulties and challenges in accessing it, as it can be segmented by different criteria (geography, age, sex, interests, etc.), which makes it difficult to trace the true expenses.

This will be a major challenge in any country experimenting with direct democracy mechanisms, whether referendums or citizens’ law initiatives, because they potentially bring in a different mix of actors than those seen in traditional elections.

10.5. Conclusions and lessons learned

Chile has made an exceptional effort to adapt and modernize its regulations on political campaigning in the last few years, in response to various scandals—something that is rare in the region. Also, the Electoral Service has proven to be very effective in organizing almost flawless elections and has good levels of public confidence, which is remarkable for Chile. This provides the country with a very good opportunity to innovate, and to address challenges in an area that is dynamic but requires urgent modernization.

While digital campaigning has enabled political contestants to reach voters at lower cost, the lack of sufficient regulation can actually undo its very purpose, which is to level the playing field and promote greater fairness in electoral competition. The Electoral Service in Chile, after first attempting to forbid digital advertising, tried to treat social networks and digital media as analogous to traditional political advertisements on radio and in newspapers, without considering their special characteristics. This has resulted in legislative loopholes, as well as making it challenging for regulators to enforce electoral rules. This calls for the need for better clarity in the regulatory definition of digital propaganda—for example, defining the act of paying for promotion in Google’s search engine as propaganda—and of the boundary line with freedom of expression.

These loopholes can and do lead to the following electoral integrity challenges:

• violation of permitted campaigning periods;
• use of bots for manipulation of the electorate or negative campaigning;
• misinformation campaigns;
• external influence in national elections;
• violation of rules on expenses and donations (origins, limits);
• difficulty in tracing electoral finance (‘following the money’); and
• unregistered, unauthorized campaigning, including indirect campaigning.

Domestic actors aside, it remains a major challenge to enforce local statutes on large digital companies that, although legally incorporated abroad, operate on national territory and have a significant impact on the national political scene. As Zamora (2020: 16) states: ‘Considering the law and the interpretation made by the electoral body, we would dare to suggest that the Chilean regulation is confusing, anachronistic and meagre with respect to the treatment of new technologies.’

10.6. Next steps

One proposal made by the Engel Commission but not considered in new regulations was a public register of suppliers. This would promote more competence on standardized services, ensure better oversight of electoral expenses and fairer pricing, and prevent conflicts of interest or even possible overpricing or abuse of public resources (currently, Chile is seeing ongoing investigations into possible fraud) (Castañón 2023; Engel Commission 2015: 81; Ayala 2023). Although the Engel Commission proposal goes beyond the difficulties mentioned here, it could be a useful mechanism to address some of them, such as providing better information to the Electoral Service on which companies political parties or candidates have engaged with for digital campaigning, the services they have provided, and the amount of resources destined for digital and online campaigning.

As is already the case with professional and volunteer campaigners (‘brigadists’), candidates should also be responsible for and have civil liability for the any wrongdoing of suppliers or companies that provide electoral advertisement services. Latin American media institutions have reported how irregularities by such suppliers are more common than expected and are transnational in nature (Fundación LaBot 2023). Also, collaboration with civil society organizations to report breaches of regulations on social networks could help in complementary monitoring.

Electoral stakeholders in Chile should consider the following:

• Regulate electoral propaganda within social networks and digital media expressly and specifically. Reporting requirements and permitted timelines for campaigns must take account of specific dynamics, such as the hiring of bots, interactions with influencers and supporters, grassroot campaigns, foreign funding. Transparency on this issue is key to promoting compliance with existing rules.
• Strengthen the capacity of the Electoral Service, especially, but not only, during electoral periods. Additional financial and human resources are needed for effective oversight and to ensure the continuous innovation that is needed to keep up with the fast pace of developments in the digital campaign realm.
• Collaborate with counterparts and expert partners regionally and globally. As regulating big tech is ultimately beyond the capacity of individual countries and their oversight agencies, Chile should seize opportunities to pool resources with others, including through the European Union and other partners. Latin America needs to confront digital capacity building and regulation in the light of other transnational problems including organized crime.
• Implement current best practices as an interim measure. As an example, Google has special regulations on political advertising in some countries, but not in Chile, so there is scope for negotiating improvements in the short term before more ambitious solutions can be identified.
• Adopt a whole-of-society approach. It is not only big tech that needs to be called to collaborate better, but also national political parties. The Electoral Service can play a role in bringing political forces together, so that they commit to not only comply with political finance laws but also follow basic standards of ethical political campaigning, including refraining from spreading misinformation through platforms, committing to high standards of transparency and reporting, and working with companies that carry out digital campaigns ethically. Finally, monitoring of online campaign finance goes beyond the realm of the oversight body and should involve other actors such as data protection bodies, commissions on digitalization, etc.

Wouter Wolfs

11.1. Introduction

The EU is the most developed supranational political community in the world. Since its inception, the EU member states have gradually transferred more and more competences to the European level. Yet, despite the EU’s growing responsibilities, the regulation of politics and elections—and particularly its financial aspects—has long remained the exclusive domain of national authorities. However, recent years have seen a remarkable shift in the EU’s position. This has followed controversies like the Cambridge Analytica scandal—in which the personal data of millions of voters were collected and used for personalized targeted political campaigning—and (alleged) Russian interference in several United States and European elections. The EU has taken several steps that have substantially increased its regulatory reach over online—and even offline—political advertising.

Just like the EU itself, European politics is a multi-level phenomenon, with separate but connected political parties existing at both the EU and the national level. Political parties at the European level—also called ‘Europarties’—function as umbrella organizations, bringing together national parties with a similar ideological background. For example, the European People’s Party (EPP) comprises 50 Christian democratic and centre-right national political parties. Currently, there are 10 such European political parties (see Table 11.1). Overall, they are rather loose associations: they develop common manifestos, select a campaign figurehead for the European elections and hold coordination meetings for their national party leaders and (prime) ministers, but they rarely impose political positions in a top-down manner.

In 2004, the EU established public funding for European political parties, together with a set of rules on how these financial resources may or may not be used (Wolfs 2022). This regulatory framework only applied to parties at the European level and had few consequences for political finance at the national level, which continued to be regulated by national provisions. It is only more recently that the European level has introduced new legislative acts with important implications for party and campaign finances—in particular, for its digital and online aspects.

This case study will examine the implications of these new EU rules for political finance, with a specific focus on online advertising. The second part of the study provides an overview of the regulatory framework governing the financing of political parties and electoral campaigns. The third part focuses specifically on the European political parties’ expenditure on social media platforms. In the fourth part, the latest EU-level regulatory efforts will be examined in detail, particularly the rules on online advertising. The case study ends with a discussion of the best practices and challenges that emerge from the new regulatory framework.

11.2. A multi-level regulatory framework

In the multi-level political system of the EU, the regulation of political finance is a competence that is shared by the authorities at European and national level. Rules applied at one level of government have an effect on political competition at the other level. Lawmakers at both levels have faced an important choice. One option is to ‘mainstream’ digital and online aspects, incorporating them into the existing rules on political finance and electoral expenditure. Another option is to address them through tailor-made legislation that focuses specifically on the online organization and campaign activities of parties and candidates. At both European and national level, a combination of both approaches can be observed.

11.2.1. Political finance regulation at the national level

The finances of national political parties are regulated by the 27 national authorities and sometimes even by regional authorities. Because regulatory traditions vary widely across countries, this has resulted in a plethora of different rules across the European continent. For example, the EU member states differ in terms of donation ceilings, electoral periods, campaign expenditure caps and the regulation of third-party campaigning (see, for example, European Parliament 2021). Specifically with regard to the digital aspects of political finance, there is also substantial variation in the online availability and accessibility of information on the revenue and expenditure of political parties and candidates (Wolfs 2023).

Similarly, EU member states have diverse approaches to online campaigning. Some countries, such as Sweden, do not have any rules in place regarding electoral advertising. Others, like Denmark and Italy, only regulate political advertising on selected media, such as radio and television, and do not usually extend these rules to online political advertisements. Many countries have a wide and generic regulatory framework for electoral campaigning, in which some or all of the provisions also apply to the online sphere. This is the case in Belgium, Croatia, Czechia, Estonia and Portugal, for example. A limited number of EU member states, including Ireland and France, have developed specific rules for online campaigns and political advertising on social media (Heinmaa 2023). In 2016, Latvia introduced a specific section on online campaigning into its campaign finance regulation which focused on transparent pricing and sponsorship of online advertisements (Cigane 2022).

These different national approaches have important implications for political parties at the European level. In several EU countries, electoral campaigning is limited to (national) political parties or candidates, which means that Europarties cannot conduct campaign activities. For example, in Slovenia, only parties (or citizens) with their seat in the country can campaign. Only four countries—Germany, Ireland, Latvia and Romania—explicitly allow political campaigns by European political parties.

In several countries, some campaign activities by Europarties may be considered as in-kind donations to national parties or Europarties may fall within the definition of ‘third parties’ and thus have to comply with the applicable rules. In Czechia, a third party may not be a foreign legal entity, as is the case with European political parties (Reed 2023). In Belgium, the campaign expenses of third parties—like Europarties—are considered to form part of the national parties’ expenses for the purposes of assessing whether they have respected the expenditure threshold, and European parties must also respect third-party campaign limitations, such as bans on radio or television advertising or large billboards (see also Vanden Eynde 2023).

Other aspects of the regulatory framework can also constrain Europarties’ campaign activities. In France, all commercial advertising through the press or audiovisual channels—including online campaigning—is prohibited in the six months leading up to an election. These substantial differences in the regulation of campaign finance across member states have prompted one Europarty official to describe political campaigning in European elections as ‘playing football simultaneously on 27 different soccer fields with 27 different sets of rules’ (Wolfs 2022: 233).

11.2.2. Political finance regulation at the EU level

The EU’s 2004 regulatory framework specifically addresses the status and funding of political parties at the European level. In general, the legislative acts that regulate the organization and activities of these European political parties conform to the ‘traditional’ approach: there are no specific provisions on the digital aspects of politics and campaigning, but some of the rules have effects beyond the classic aspects of party finance. The framework also has some implications for national political parties, albeit relatively few. For example, as Europarties cannot financially support national parties or candidates, either directly or indirectly, it prohibits financial transactions, contributions in kind or campaigning on behalf of these parties or candidates. Overall, however, these provisions are targeted at the Europarties, with only limited direct effects on the political competition at national level.

The situation started to change in 2019, when the EU initiated several regulatory steps to confine campaigning—and particularly digital campaigning—as part of a broader effort to restrict online advertising. The main catalyst for legislative action has been the coverage of the Cambridge Analytica scandal. This involved the use of high volumes of personal data to engage in profiling and targeted political advertising on social media, and it was linked to possible foreign interference in elections and the Brexit referendum (European Commission 2018a, 2018b, 2019; European Parliament 2018, 2019). The first attempts at regulation relied on soft law, but after these proved insufficient, the EU institutions have since issued binding legislative acts.

The EU initially focused on self-regulation to tackle the challenges of online advertising. In 2018, the European Commission launched a Code of Practice on Disinformation, a voluntary agreement, signed by the major online platforms such as Meta (Facebook), Google, Mozilla and X (formally Twitter) (European Commission 2018c), with Microsoft and TikTok joining in 2019 and 2020 respectively. With regard to political campaigning, the code included a commitment by the signatories to clearly label political advertisements as such and provide information about the sponsors and the amounts spent. In 2020, a European Commission assessment identified a number of shortcomings (European Commission 2020) and a new version of the Code was launched in 2021 (European Commission 2021, 2022). In this ‘strengthened’ version, the online platforms, among others, committed to (a) improving the labelling of political advertising; (b) identifying the sponsors of advertisements through a verification procedure; and (c) maintaining a repository of all political advertisements. Overall, the European Commission’s main effort in introducing this self-regulatory approach was to substantially improve the transparency of online political campaigning.

Over time, however, the EU institutions developed a binding regulatory framework through multiple legislative initiatives. This was partly because the results of the Code of Practice remained insufficient. The new rules dealt with two main issues (see also Gibson, Bon and Römmele 2023): (a) the regulation and restriction of access to (large quantities of individualized) data for use in political advertising, their collection and analysis, and their application in political communication; and (b) improved transparency and reporting of sponsored political advertisements, both online and offline.

Regarding data access and collection specifically, the first legislative initiative to be taken was the General Data Protection Regulation (GDPR) in 2016 (Regulation 2016/679). This new law centred around standardizing and restricting the use of European citizens’ personal data and limiting the scope for political parties and candidates to use these data for the purpose of political advertising. In 2019, the rules on European political parties were updated to bring them in line with the GDPR (Regulation 2019/493). More specifically, a provision was added to emphasize that European political parties are forbidden from influencing, or attempting to influence, the outcome of the European elections by infringing the rules on the protection of personal data. Currently, the EU institutions are working on new rules to complement the GDPR. The Regulation on Privacy and Electronic Communications, or ePrivacy Regulation, is still under negotiation but aims to impose stricter limitations on the ability of organizations to collect data about the online activities of citizens (European Commission 2017). For example, it abolishes the provision that prior consent was not required in cases of overriding public interest.

In 2022, the EU institutions agreed on a new law to further strengthen the protection of fundamental rights in the online sphere—the Digital Services Act (DSA). An important part of it was dedicated to online advertising and introduced several measures to improve transparency. While the DSA established rules that apply to all online advertisements, the EU also developed specific rules for political advertisements. As part of its European Democracy Action Plan in 2020, the European Commission announced a new initiative to improve the transparency of political advertisements. In 2023, a political agreement was reached on this new Regulation on Transparency and Targeting of Political Advertising. This piece of legislation breaks with the traditions of most member states. For the first time, specific rules were developed regarding online campaigning, with far-reaching implications for traditional physical political campaigns, at both the European and the national level. (For a more detailed discussion of this regulatory framework as a whole, see 4: Key features of the regulatory framework.)

11.3. The use of social media advertising by European political parties

Overall, the social media campaign expenditure of European political parties has been relatively modest, especially considering they operate on a continent-wide scale. Table 11.2 provides an overview of the social media expenditure of the 10 registered Europarties from 2019 to 2023. Except for the run-up to the 2019 European elections, the annual total expenditure of all Europarties combined has been around or lower than EUR 100,000. But even during the electoral period of 2019, the social media campaign expenses of the European political parties were relatively low. The only party to spend a substantial amount was the European Green Party, which devoted more than EUR 400,000 to social media in 19 EU countries—around a quarter of its total campaign budget. This was still relatively modest in comparison to the social media expenditure of more than EUR 3.2 million by the European Parliament, which conducted a ‘Get out to vote’ campaign. None of the other parties conducted a large social media campaign.

This limited social media expenditure can be largely explained by the regulatory framework that was in place in 2019. In the run-up to the elections, Facebook required all advertisers to physically register in the country where they wished to purchase political advertising. For most European political parties, which had their registered headquarters in Brussels, this meant that they could only campaign digitally in Belgium and were prevented from using social media to conduct genuine transnational campaigns across all EU countries. The use of this measure by Facebook was the result of the soft law approach towards online campaigning that was being taken at the EU level at the time. Online platforms could develop their own policy on political advertising, and, according to Facebook, this was the best way to tackle foreign interference. Yet, in imposing this limitation, the online platform ignored the transnational nature of European political parties, which, by definition, operate and campaign across national borders. It was only after substantial pressure from the three EU institutions—and just one month before the elections—that Facebook agreed to make an exception for the European political parties (Wolfs and Veldhuis 2023).

11.4. Key features of the regulatory framework

11.4.1. Limitations on data access and data analysis

An important goal of the regulatory efforts at the EU level has been to curb the unlimited collection of personal data for micro-targeting voters through online political advertisements. Such practices were first regulated by the GDPR in 2018, which requires organizations to obtain citizens’ consent for the collection and processing of personal data. However, the GDPR allows the collection and use of these data without consent if it is necessary for the performance of a task carried out in the public interest or if it is necessary for the purpose of legitimate interests of the party or candidate (article 6(e) and (f) of Regulation 2016/679).

In addition, the GDPR bans the use of ‘sensitive’ personal data, which include citizens’ racial or ethnic origin, political opinions, religious or philosophical beliefs, trade union membership, genetic or biometric data, and information about their health or sexual orientation. However, there are many exceptions. For example, processing is allowed if a citizen provides explicit consent for one or more specific purposes or, in analogy to ordinary personal data, in cases where it is necessary ‘for reasons of substantial public interest’ ( article 9(2)(a) and (g) of Regulation 2016/679). In this respect, the preliminary argumentation of the rules stipulates that the compilation of personal data on voters’ political opinions by parties in the course of their electoral activities may be permitted for such reasons of public interest, provided that ‘appropriate safeguards are established’ (recital 56 of Regulation 2016/679). Parties can also process these data if they originate from members or persons who have ‘regular’ contact with the party (article 9(d) of Regulation 2016/679), such as their members.

The GDPR also contains important provisions on automated processing of personal data such as profiling, which can be used to analyse and predict voting behaviour. This can be a powerful tool for political parties and candidates because it can be used to micro-target voters and thereby maximize the impact on their voting behaviour. Automated profiling and automated targeting are also prerequisites for the kind of large-scale political micro-targeting exemplified by the Cambridge Analytica scandal (Brkan 2022: 357).

The GDPR stipulates that citizens ‘have the right not to be subject’ to such automated processing techniques, at least to the extent that such techniques have legal effects or significantly affect the circumstances, behaviour or choices of the citizens concerned (article 22 of Regulation 2016/679). There is some debate over the extent to which (micro-)targeted political advertising has such effects and, consequently, whether this provision actually prevents political parties from using profiling or similar techniques in their online campaign activities (Article 29 Data Protection Working Party 2017; Brkan 2022: 357–59). Yet there are strong arguments to support such an assessment. Since automated and (micro-)targeted campaigning hampers the unrestricted flow of political ideas and voters’ access to all sources of information, it can contribute to the spread of disinformation. This, in turn, threatens pluralist political debate and the free formation of thoughts and opinions. The risk is heightened by the fact that targeted political advertising requires campaign funds—sometimes substantial sums—and thus favours parties with a larger campaign budget over those with fewer financial resources (Dobber, Ó Fathaigh and Zuiderveen Borgesius 2019; Levano 2023). There is, in other words, a substantial effect on the behaviour or choices of individuals.

Despite the ambitions of these EU-level regulatory efforts, several EU member states have introduced exemptions for political parties in their implementing acts of the GDPR. Although the latter is an EU regulation, which is directly applicable and does not require transposition into national legislation, one third of the articles of the law permit member states to deviate from the provisions as stipulated in the EU legal text (Bender 2018). This is also the case for provisions relating to the processing of (sensitive) personal data. For example, the Romanian data protection law included a provision allowing political parties to collect and process sensitive personal data without the explicit consent of voters (article 9(2) of Law No. 190/2018) (Romania 2018). In Spain’s implementation of the GDPR, a specific provision was added to the electoral law that permitted political parties to process personal data from publicly available sources and collect sensitive personal data on political opinions without consent in the framework of their electoral activities (article 58bis of the Organic Law No. 5/1985 of the General Electoral Regime, as modified by Organic Law No. 3/2018) (Spain 2018). However, after a complaint from the Ombudsman, the provision was annulled by the Spanish Constitutional Court (Judgement 76/2019 of 22 May 2019). In addition, in the United Kingdom—which was part of the EU until 2020—political parties were and are allowed to process personal data on citizens’ political opinions without the need for consent for use in their political campaigns, fundraising, political surveys and casework (article 9(2)(g) of the UK GDPR; paragraph 22 of schedule 1 of the 2018 Data Protection Act).

In sum, while the GDPR was aimed at harmonizing the use and processing of citizens’ personal data—including by political parties—the rules and implementing measures developed at national level still allow for a high level of variation between the member states. There also remain a large number of loopholes allowing political parties to use (sensitive) personal data in their political campaigns.

With regard to European political parties in particular, the EU strengthened its regulatory provisions on the protection of personal data in response to the Cambridge Analytica revelations in 2018. This was deemed particularly urgent in view of the European elections taking place in May 2019. The Regulation on the Status and Funding of European Political Parties was amended specifically in order that European political parties could not ‘deliberately influence, or attempt to influence, the outcome of elections to the European Parliament by taking advantage of an infringement by a natural or legal person of the applicable rules on the protection of personal data’ (article 1(3) of Regulation 2019/493). This would, for example, be considered to have occurred if a European political party were to engage in micro-targeting of political advertising without explicit consent of the voters concerned (see e.g. Brkan 2022: 360). Where a European party is found to have breached these rules, it can receive a financial sanction of up to 20 per cent of its annual budget (in the case of a repeated infringement) (point 6 of article 1 of Regulation 2019/493).

The rules also aim to forestall a situation whereby a European political party buys services or information from a data broker or a data analytics company that infringe the data protection rules, in analogy to the Cambridge Analytica scandal. The data broker or analytics company—or any other natural or legal person concerned—can only be sanctioned by a national supervisory authority in the case of an infringement; European political parties, however, can be investigated by the EU monitoring entity—the Authority for European Political Parties and European Political Foundations—if they are found to be relying on such breached data or if there are reasonable grounds to believe that the infringement is linked to its political activities (article 1(3) of Regulation 2019/493).

In 2022 and 2023, to address the loopholes that the GDPR had kept in place, the rules were tightened on the collection and processing of personal data with the purpose of using these for (political) advertising. The Digital Services Act imposes transparency requirements for data processing and targeting techniques, as well as banning the use of targeting and amplification if it involves automated processing of sensitive personal data, such as profiling. The Regulation on the Transparency and Targeting of Political Advertising takes the requirements a step further by allowing targeting or similar ad delivery techniques only if (a) these techniques do not involve sensitive personal data; (b) the party or candidate has collected the data from citizens; and (c) citizens have given explicit consent for their personal data to be processed. Targeting of young people below the voting age is entirely prohibited. These restrictions do not prevent general communication from parties to their (former) members, such as newsletters (article 18 of forthcoming Regulation, see European Parliament 2024). Consequently, these new rules substantially limit the scope for political parties to make use of sensitive personal data, such as political opinions, in their online campaign strategies.

11.4.2. Strengthening of transparency

While the GDPR was mainly aimed at providing a framework—and limitations—for the collection and use of personal data, the most recent legislative initiatives have also focused on improving the transparency of political advertising. The Digital Services Act stipulates that advertisements on online platforms should be clearly labelled as such to allow users to differentiate them from other content. In addition, an advertisement must identify both the natural or legal person that paid for it and the natural or legal person on whose behalf it is presented (article 26 of Regulation 2022/2065). Applied to political advertising, this means that both the sponsor of a political advertisement and the party or candidate that is promoted must be made clear. Consequently, these provisions also entail de facto transparency in online campaigning by third parties. The DSA sheds light on targeting, since users must have easy access to the parameters that were used to determine the recipients of an ad. In addition, the law explicitly prohibits the targeting of advertising using profiling on the basis of sensitive personal data (article 26 of Regulation 2022/2065).

The DSA strengthens the positions not only of the recipients of advertisements but also of monitoring bodies. It thus offers the potential to strengthen the monitoring and oversight of online political advertising. According to the rules, ‘very large online platforms’—those with more than 45 million users per month, such as Facebook, Google, Instagram, TikTok and YouTube—must keep a repository of all published advertisements. The repository should include the content of the advertisement, the person or entity on whose behalf it was presented, the donor, the period during which the advertisement was published, the parameters used to target specific groups of users, and information on the total number and groups of recipients (article 39 of Regulation 2022/2065).

While the DSA is applicable to all advertisements, the EU has also developed a specific regulatory framework for political advertisements—the Regulation on Transparency and Targeting of Political Advertising—with the aim of fostering an open and fair debate in the run-up to elections (recitals 4, 8, 19 and 64 of forthcoming Regulation, see European Parliament 2024). The regulation aims to achieve this by obliging parties and other political actors to make it clear to citizens when they are exposed to (paid) political content.

While the regulation has a specific focus, it is also wide-ranging in scope. Particularly noteworthy is its broad definition of ‘political advertising’, which goes beyond what is applicable in most EU member states:

the preparation, placement, promotion, publication, delivery or dissemination, by any means, of a message, normally provided for remuneration or through in-house activities or as part of a political advertising campaign by, for or on behalf of a political actor …, or which is liable and designed to influence the outcome of an election or referendum, voting behaviour or a legislative or regulatory process, at Union, national, regional or local level.

(article 3(2) of forthcoming Regulation 2024, see European Parliament 2024)

First, ‘the preparation … or dissemination, by any means’ implies that the rules encompass both online and offline political advertisements, including political posters, pamphlets, newspaper advertisements, and radio and television commercials as well as social media advertisements. Second, the rules are applicable not only to the advertising activities of political parties and candidates but to all individuals and organizations engaged in political advertising. In other words, all forms of third-party campaigning—both online and offline—fall within the scope of the rules. Third, the legislative act applies to political advertising during as well as outside of elections and entails all sponsored content intended to influence political decision-making processes. In many countries, specific rules are in place for political campaigning during electoral periods, but the new EU rules affect all forms of sponsored political communication, including those that take place outside of elections.

The regulation requires the labelling of every political advertisement with a ‘transparency notice’. This should include the identity and contact details of the sponsor; the period of publication; the amounts spent and the source of funds; and the election or referendum to which the advertisement is related. In addition, specifically for online advertisements, it must be made clear when an advertisement is targeted, what criteria have been used for that targeting, and whether and to what extent amplification techniques were used to boost the advertisement’s reach. If a political advertisement does not include this information, it cannot be published.

The regulation also established an important source of information for political finance observers and monitoring entities. ‘Very large online platforms’ as defined by the DSA—such as Facebook, Google, YouTube, Instagram, TikTok and X (formally Twitter)—must keep a database of all online political advertisements, including all the data required in the transparency notice, including sponsors and amounts paid. All providers of political advertisements, both online and offline, must keep these data for a period of seven years. Consequently, the European rules have created a situation in which almost all sponsored political communication is registered and stored. This has provided a unique opportunity for political observers to map trends and analyse spending patterns. It also enables oversight and audit bodies at both the European and the national level to verify these data against the financial reports of political parties and candidates.

11.5. A multi-level monitoring field

The multitude of new European-level legislative initiatives have created new opportunities for observers and monitoring entities. Yet they also present a significant challenge, since they imply that responsibility for the monitoring and enforcement of the rules is dispersed among several governmental bodies.

The GDPR foresees the establishment of national data protection authorities to verify compliance with the data protection rules (point 21 of article 4 of Regulation 2016/679). However, in the context of European elections, in cases where a breach of these rules is linked to the political activities of a European political party, the national supervisory authority must inform the Authority for European Political Parties and European Political Foundations. The Authority must request an opinion from a committee of independent eminent persons and take this opinion into consideration when deciding whether or not to impose a financial sanction (point 3 of article 1 of Regulation 2019/493). In other words, the monitoring entity at EU level—the Authority—is entirely dependent on the decision of the national supervisor to start its own proceedings.

Similarly, the DSA divides the task of oversight between (a) Digital Services Coordinators and other competent national authorities and (b) the European Commission, which is responsible for monitoring very large online platforms. The Regulation on Transparency and Targeting of Political Advertising again places oversight responsibility at the national level, requiring the establishment of new entities with responsibility for monitoring its implementation. Only they can impose administrative fines and financial sanctions. However, in order to maintain an effective monitoring system, all these entities on multiple levels must be able to exchange information and collaborate. If we consider that the key challenge for political finance is the enforcement of the existing rules, it immediately becomes clear that the success of the new regulatory framework depends on meeting this same challenge.

11.6. Lessons learned and future prospects

Through its legislative initiatives of recent years, the EU is increasingly entering a policy field that has traditionally been reserved for the national level—the regulation of political parties and campaign finance. In this respect, the European level has chosen to focus on the online environment and has imposed substantial limitations on the ability of political actors to collect personal data and use these in their political campaigns. In addition, the disclosure requirements included in the new rules hold the potential to usher in an unprecedented era of transparency in campaign expenditure—not only from political parties and candidates but from all actors seeking involvement in sponsored campaigns and influence over electoral or decision-making processes. While these new regulations offer important opportunities for civil society and monitoring entities to scrutinize the funding of political campaigns, they may also pose coordination challenges, given that monitoring and enforcement competence is dispersed among many actors across multiple countries and governmental levels.

While the new regulatory framework offers important improvements to the use of online advertisements in political campaigns, its effectiveness will depend on its implementation. Several online platforms have already started to update the transparency requirements they place on political advertisements, sometimes going further than the current rules require. For example, Meta requires advertisers to disclose the use of artificial intelligence or other digital techniques to create or change a political advertisement (Clegg 2023). Yet the first elections organized after the DSA’s entry into force were characterized by a fight against disinformation campaigns, which shows that compliance remains challenging (see e.g. Hartmann 2023; Scott 2023). Similar compliance challenges are to be expected with regard to the transparency requirements and data collection and processing rules. In addition, it should be noted that around 10 per cent of all online political advertisements are not published on the so-called ‘very large online platforms’ and thus fall outside of the scope of the transparency rules (European Partnership for Democracy 2022).

Nevertheless, the EU case provides a number of best practices that can serve as an example to other countries:

1. In our age of digital campaigning, it is often not sufficient to regulate only the behaviour of political parties and candidates. (Social) media platforms also need to be included in the legal framework and integrated into the broader chain of enforcement of the rules.
2. While the use of soft laws, such as codes of practice, can be an important first step, this is often an insufficient means of providing an impermeable legal framework. Indeed, giving online platforms (or political parties) substantial discretion in designing and implementing such initiatives can lead to possible unintended consequences.
3. Adopting a broad definition of political advertising ensures that all actors that attempt to influence the electoral and decision-making process are regulated. This includes influencers and third parties that are active in the political campaigns.
4. The focus on increasing transparency can be a feasible policy goal in political systems with widely diverging positions or practices regarding the regulation of political finance. The EU and its member states are characterized by a wide variety in traditions of political finance and the competences of the European level are limited in this regard. Nevertheless, it has been possible to find a consensus around bolstering transparency of political advertisements and the use of personal data with the common goal of strengthening the EU’s democratic foundations.

Niranjan Sahoo

12.1. Introduction

With a population of over 1.4 billion and 986 million voters, India is the world’s largest democracy. It is a multiparty, federal democracy with 6 national parties, 54 state/regional parties and over 2,000 unrecognized political parties (see Rathore 2023). While 13 states are currently run by regional parties, two national parties—the Bharatiya Janata Party (BJP) and the Indian National Congress (INC)—between them govern in 15 Indian states. The BJP, which has been in power at the federal level since 2014, is currently in government across 17 Indian states, governing alone in 12 and in coalition in 5 (for a detailed view, see Hindustan Times 2024).

Since the BJP’s historic electoral sweep in 2014, under the leadership of Prime Minister Narendra Modi, India’s electoral scene has witnessed a rapid escalation in election campaign spending (see Figure 12.1). It is noteworthy that most parties, particularly the national parties such as the BJP and the INC, allot a substantial proportion of their party funds to publicity/election advertisements (see Figures 12.2 and 12.3). India has experienced a huge growth in social media users in the past decade, and political parties and candidates are increasingly deploying social media platforms in their electioneering, in particular to reach out to younger voters.

12.2. Digital campaigns: Key trends

Elections in India are aptly called ‘Festival of Democracy’. Once the poll date has been announced, a carnival-like atmosphere sets in where political parties, candidates and their supporters can be seen campaigning door-to-door to the sound of drums, loud music, traditional recitals and chanted slogans. While the traditional mode of campaigning still rules the roost, a considerable amount has shifted online or to the digital sphere. The most significant use of digital technology in election campaigning was during the 2014 general election. This was the first election to see major political parties deploy big data analytics to crunch information on millions of new social media users to assist their campaigns. Furthermore, political parties not only hired public relations executives and technical consultants, but proactively reached out to voters on Facebook, Twitter, TikTok and YouTube, among others. According to one reliable estimate, political parties possibly spent around INR 500 crore (USD 82.91 million)¹⁷ on dominating the social and digital media space during the 2014 polls (Dutta 2014). Although social media was mostly confined to the young, urban middle class, these sections are the real opinion makers, so reaching them created maximum buzz during the 2014 general election. The BJP and its candidate, Narendra Modi, launched the most innovative and aggressive digital media campaign that dominated most social media platforms (Rao 2019). Analysts felt that Modi’s innovative social media outreach contributed greatly to his success in the 2014 election (Sen, Naumann and Murali 2019). Modi’s spectacular electoral victory smartly harnessed the potential of digital media, prompting the London Financial Times (Chilkoti 2014) to call him the ‘first social media prime minister’.

However, it was the 2019 general election, which was the most expensive election ever, that can truly be called a watershed election as far as social media is concerned. According to some estimates, political parties and candidates spent USD 8.7 billion on the 2019 general election (Pradhan and Kumaresan 2019). There had been massive growth in social media users in a country where more than half the population was under the age of 30 in 2019 (Indian Express 2022). India’s Internet users had reached 560 million by January 2019, with 310 million social media users and a whopping 1.19 billion active cellular mobile connections (Kemp 2019).¹⁸

Aided by cheap Internet data and the use of budget smartphones (second only to China), the digital divide as witnessed in the 2014 polls had considerably narrowed. Social media platforms such as Facebook, Google, YouTube, WhatsApp and Instagram dominated political communications and narrative-building, and political parties spared no efforts to dominate the new and potentially transformative medium for political communications. Staying ahead of the curve, the BJP, which had created its famed ‘IT Cell’ for the 2014 general election, deployed more than 1.2 million social media volunteers and thousands of troll armies to dominate the narrative and push the party’s electoral agenda. The NaMo app, which captured all the major activities of Prime Minister Modi, was a standout feature of the ruling party’s social media campaign strategy in the 2019 polls (Sen, Naumann and Murali 2019).

While the BJP retained the ‘early mover’ advantage, nearly all the political parties, particularly the main opposition the INC, gave a major push to improve their digital media capabilities during the campaign. According to media analysts, the INC spent more than INR 200 crore (USD 32.27 million) on social media campaigning, around 10 times its 2014 social media campaign budget, to counter the BJP-led government’s digital platforms (Roy 2019). The party attempted to re-energize its social media outreach by establishing a dedicated social media team to stay afloat in the game. It trained social media staff, proxy workers and hundreds of volunteers to generate innovative digital content and filter creative messaging. Many regional parties, particularly the All India Trinamool Congress (TMC), the Telugu Desam Party (TDP), Yuvajana Sramika Rythu Congress Party (YSR), Biju Janata Dal (BJD) and Aam Aadmi Party (AAP), also invested heavily to boost their social media outreach among voters (Udupa 2019).

Since the 2019 elections, the appeal of digital media to candidates and parties has continued to grow. More than half the population now has access to smartphones and social media channels. As more and more rural populations consume digital data, political parties are going all out to overhaul their social media campaign strategies and infrastructure. This has been vividly demonstrated in state polls, where regional and smaller political parties compete with national parties to leverage the potential of digital media (Shekhar 2022).

A striking feature of political communication through social media in India is the rapid expansion of social media applications in regional languages. By 2019, most political parties were using regional language platforms to enhance their support among electorates. Two standout social media platforms that emerged were ShareChat and Helo, which catered to regional languages, operating in as many as 15 different languages, including Hindi, Tamil, Telugu and Bengali. In no time, ShareChat was able to target 100–150 million mobile Internet users in rural India and Tier 2 and 3 cities (Sen, Naumann and Murali 2019). Thus, a major transformation had taken place between 2014 and 2019 as far as social media communications were concerned.

12.3. Digital campaigning expenditure in numbers

As discussed above, the massive jump in the number of social media users in the past decade has led political parties both large and small to make huge efforts to leverage the potential of digital media in election campaigns, as well as political and related outreach. One point that needs to be emphasized, however, is that while regional parties have greatly improved their presence in social media and campaign spending, digital media election campaigns are largely dominated by the two national parties, and the BJP has the lion’s share.

The BJP has invested heavily in its digital infrastructure and personnel to maximize its reach on social media (Mehta 2019). An analysis of political advertisements placed on the two major digital platforms, Google and Facebook, which unlike other digital platforms launched advertisement transparency initiatives ahead of the 2019 election, shows that the BJP had overwhelming dominance in social media spending. Cumulative digital advertising spending by political parties on Google and Facebook during the campaign period (February to May 2019) was INR 58 crore (USD 8.3 million). Compared to the volume of spending on publicity (see Figure 12.3) by most political parties, this sum appears minuscule. However, this might be because these figures do not include expenses in connection with content development, salaries/fees for social media personnel, and so on (Chaturvedi 2019). Thus, what is visible in the public domain is a fraction of total spending by parties and candidates.

The BJP topped the list of social media spending by key parties. Of the INR 28.4 crore (USD 4 million) that Facebook received for advertising, the BJP spent INR 4.3 crore (USD 615,750), while the INC spent around INR 1.8 crore (USD 257,760) (see Table 12.1). Among the regional parties, All India Trinamool Congress (TMC) spent INR 29.28 lacs (USD 41,930) during the 2019 elections. Political advertising on Google varied slightly. While INR 29 crore (USD 4.15 million) was spent on Google advertising and its associated platforms, the BJP alone spent 60 per cent of the total. The second highest spender was the Telugu Desam Party (TDP), a regional party based in Andhra Pradesh and Telangana. The third top spender on Google was Dravid Munnetra Kazhagam (DMK), a regional party based in Tamil Nadu. The INC spent just INR 2.71 crore (USD 388,070) on Google for the 2019 polls.

Since 2019, political advertising on social media has taken an upward turn, particularly as more regional parties seek to enhance their social media strategies to match the ruling BJP. A major explanation for this is the Covid-19 pandemic and the associated restrictions imposed by the Election Commission on political rallies and door-to-door campaigning. For instance, on the 2021 elections for the five states of Uttar Pradesh, Goa, Uttarakhand, Manipur and Punjab, political parties spent INR 12.6 crore (USD 1.7 million) on Facebook (Rampal 2022). In the recently concluded elections for the five states of Madhya Pradesh, Rajasthan, Chhattisgarh, Telangana and Mizoram, in September to November 2023, there was a substantial increase in political advertising on social media. As per data collected by the Association for Democratic Reforms (ADR), INR 53.5 crore (USD 6.42 million) was spent on advertising by political parties on various social media platforms. A much greater acceleration in social media spending by political parties is expected in the upcoming general election in early 2024. A large youthful population is literally conversing on social media (CSDS 2019). Content moderation, the challenges of disinformation and hate messages, and regulation of political advertisements/endorsements by third parties/supporters will present major challenges for the ECI and other associated institutions.

12.4. Overview of existing political financial regulations

Broadly speaking, India’s campaign finance rules and operating principles are enshrined in the Representation of the People Act (RPA), 1951. However, the Election and Other Related Laws (Amendment) Act, 2003 is also important legislation. Company and individual donations to political parties are 100 per cent tax deductible under sections 80 GGB and 80 GGC of the Income Tax Act, respectively. It is mandatory for political parties to submit a list of donations over INR 20,000 (USD 240) to the Election Commission of India (Eswaran and Vaishnav 2018). Nationally, there have been no restrictions on contributions to political parties since the Finance Bill of 2017 removed the cap on corporate donations to political parties (Vaishnav 2019). Corporations were previously allowed to donate up to 7.5 per cent of average net profits over the three previous years. Furthermore, India has removed an earlier bar on foreign contributions to political parties. The Finance Bill of 2017 amended the Foreign Contribution (Regulation) Act of 2010 by inserting a clause that a company will ‘not be deemed a foreign source’ if the ‘nominal value of share capital is within the limits specified for foreign investments’. In effect, this allows Indian subsidiaries of various foreign companies to donate any amount to political parties (Srivas 2018). There is no cap on donations by individuals, other than anonymous donations which are restricted to INR 20,000 (USD 240).

In terms of election expenditure, the 2003 act puts an upper cap on candidate spending in elections for the Lok Sabha and regional legislative assemblies. This includes money spent by ‘third parties’ for a particular candidate. According to the most recent ECI direction, the upper cap for Lok Sabha elections is INR 95 lacs (USD 114,000). However, there are no limits on political party spending on election campaigns. This exemption for parties often leads to widespread violations or redundancy of the cap on candidate spending.

There is no specific legislation on expenditure on social media campaigns by parties or candidates. The ECI in partnership with social media companies has developed guidelines and a voluntary code of ethics to regulate advertising and spending on digital or social media campaigns. This apart, the ECI’s Model Code of Conduct (MCC), which remains in effect throughout the campaign period, also applies to social media campaigns. The ECI has been attempting some form of monitoring and regulation of social media expenditure since 2008. In 2013 the federal election management body rolled out elaborate guidelines for social media companies, candidates, political parties and third parties, among others (Election Commission of India 2013). The ECI defined social media as a means of interaction among people on which they create, share or exchange information and ideas in virtual communities and networks. The ECI maintains that existing legal provisions related to election campaigning should apply to social media in the same way as they apply to any other forms of election campaigning. The ECI has made it mandatory for candidates to disclose their social media accounts when they file their nominations. Significantly, the ECI instructed Media Certification and Monitoring Committees (MCMC) at the state and district levels to issue pre-certification of political advertisements for political parties and candidates carried by Internet/social media or websites. It also directed political parties and candidates to publish their expenditure on social media (Election Commission of India 2013).

Closer to the 2019 general elections, the ECI issued much more comprehensive social media guidelines that, among other things, stated that candidates were required to file affidavits on Form-26 when they filed their nominations (Election Commission of India 2019).¹⁹ The candidates must also disclose all their social media accounts. Furthermore, the ECI extended the power of the MCMC at the district and state levels to undertake pre-certification of political advertisements released to any social media or website by political parties and candidates. Notably, the ECI expected all candidates and political parties to maintain a record of all expenditure, including any expenses incurred on social media advertisements, payments made to Internet companies and websites, operational expenditure on creative content development, salaries paid to professionals/staff, and so on. However, the ECI’s guidelines were largely silent on third parties or proxy advertisers (Election Commission of India 2019).

12.5. The most recent attempt to regulate digital campaigns: The 2019 national polls and voluntary code of ethics

By 2019, the exponential growth in social media and alarming rise of misinformation had led to growing pressure on the ECI from civil society groups to devise stringent regulations on advertisements related to social media campaigns (Newsclick 2019). Independently, the major social media platforms—Twitter, Facebook, Google, ShareChat, TikTok and most importantly WhatsApp—agreed to devise and publish a ‘voluntary code of ethics’. The code, which was drafted by the Internet and Mobile Association of India (IAMAI), was submitted to the ECI on 29 March 2019. In the meantime, the ECI had set up the Sinha Committee to look into the feasibility of regulating social media advertising during election campaigns and ways to prevent its misuse.²⁰ The ECI and the IAMAI signed an agreement on the voluntary code, which entailed four key commitments from social media companies.²¹ First, social media companies agreed to conduct education and awareness campaigns, and to establish a fast-track grievance redress channel to take action on objectionable posts reported to the ECI under various electoral laws, such as section 126 of the Representation of the People Act, 1951. Second, and notably, it was agreed that social media companies would take action within three hours of reported violations of the mandatory 48-hour period of no-campaigning and advertising silence before voting. Third, they pledged to ensure pre-certification of all political advertisements published on their channels by the government’s media certification and monitoring committees, as mandated by the Supreme Court in Common Cause vs Union of India (2005). Finally, they committed to transparently report paid political advertisements and to label them accordingly (Mehta 2019).

Drawing on its 2013 guidelines, the ECI set out in its 2019 ‘Handbook for Media’ rules on third party disclosure and procedures for parties and candidates to lodge accounts of their election expenses, including on social media. The handbook defines social media expenditure as including payments made to Internet platforms/websites for advertisements, operational expenses related to the development of campaign content or maintenance of social media accounts, and salaries/wages paid to employees or professionals who undertake the tasks of maintaining these accounts (Election Commission of India 2019). During the 2022 state elections, the ECI added a new column to candidates’ election expenses returns for the submission of information on expenses incurred on digital campaigning (Economic Times 2022). In short, the Commission created a ‘dedicated point’ for capturing social media expenditure incurred by candidates and parties. To further improve regulation intended to check the misuse of social media during campaigns, the ECI launched the cVIGIL app, which allows citizens to report instances of misuse to the ECI and its officials (The Hindu 2023).

A significant aspect of regulation related to social media expenditure on election campaigns is that they have been brought under the purview of the existing MCC. The MCC is a set of guidelines agreed on by the political parties to prevent the incumbent government from misusing its power or position during elections. Once the election schedule has been declared by the ECI, no new policies or programmes can be announced by the incumbent government and the official machinery is debarred from being used for campaigning work (Daniyal 2019). Although the MCC has no legal standing, it has over the years served as a bulwark for electoral integrity and a check on the misuse of official powers during the course of election campaigns. To check the growing misuse of social media during elections, and disinformation and hate messages in particular, the ECI has authorized the Chief Electoral Officers in the states and officers at the district level to establish complaint redress mechanisms and take expedited action.

Thus, regulations and related mechanisms for controlling social media expenditure during election campaigns have evolved with rapid transformation of the digital space.

12.6. Issues and challenges

While there have been a number of initiatives, including the elaboration of a detailed set of guidelines on digital media spending by political parties and candidates, and a voluntary code for social media companies, including pre-certification by ECI officials, they at best offer only minimal deterrence against misuse or violations.

12.6.1. Limited effectiveness of the voluntary code

While the Internet companies have independently agreed to a voluntary code, there have been widespread instances of its violation. For instance, of the 510 code violations in the 2019 elections, 93 were shortlisted but only 75 were analysed by the Commission and some inferences were made. In the end, no penalties were imposed by the Commission or its affiliated bodies for violations (Sen, Naumann and Murali 2019). Similarly, 909 posts were taken down by the social media platforms during the 2019 general elections, as per the ECI, following due grievance and complaint processes. While this may look encouraging, the numbers are clearly minuscule (Mehta 2019). The 2019 polls witnessed millions of fake or misleading posts uploaded to all kinds of platforms, but in particular to Facebook, Twitter, Google and WhatsApp (Sen, Naumann and Murali 2019). Only a few hundred were taken down by these media channels. While Facebook, Google or YouTube content and advertisements can be monitored by open sources, popular and widely used platforms such as WhatsApp are impossible to regulate or monitor. In short, there are serious question marks regarding the efficacy of voluntary codes to check harmful content in social media campaigns. The extremely poor records kept on the Code of Ethics and Broadcasting Standards or on enforcing peer surveillance and industry-evolved guidelines on commercial television news are a telling reminder of the limitations of self-regulation (Seshu 2018).

12.6.2. Difficulty of monitoring third party online campaigning

There are also challenges with regard to the accurate capture and disclosure of spending on social media advertisements by candidates and political parties. Although the 2019 polls were billed as the most expensive in the world and a social media election, the accounts of the major political parties such as the BJP and the INC show only a meagre amount devoted to social media. While parties have reported a massive increase in their publicity costs (see Figure 12.3), expenditure on social media represents a minuscule proportion. This raises the issue of third party or proxy spending on digital campaigns. As analysts have pointed out, most expenditure on digital platforms is not by political parties or candidates, but by their sympathizers or groups affiliated with political parties and candidates (Mehta 2019; Udupa 2019). Given the massive penetration of social media over the past decade, it is quite plausible that a significant proportion of campaign money is being diverted to social media platforms for digital messaging, social media war rooms, social media influencers and troll factories. What is pre-certified and reported as political advertising by candidates and political parties is only a small fraction of larger political mobilization programmes on digital platforms. Even with pre-certified and declared political advertisements on social media channels, appearances can be deceptive. A lot of advertising money is routed through separate but friendly groups (Mehta 2019). The political advertisements of the BJP and INCon Google and Facebook illustrate the complexity of third-party advertisements and the challenges of identifying and accounting for them. The absence of an intermediary liability law in India is important here. The 2000 Information Technology Act is rather weak on checking the misuse of intermediaries.²² Although central government through the Ministry of Communications and Information Technology issued a framework and advice for social media platforms in 2021, which among other things allows users to register complaints, this has been of little help in curbing misuse.²³ India’s record of enforcing regulation on political finance is therefore very weak. Disclosure and transparency norms are regularly openly violated by parties and candidates with impunity and the ECI is barely empowered to take on the violations (Eswaran and Vaishnav 2018).

12.7. Conclusion and recommendations

As India prepares for the 2024 general elections, political parties and candidates are already working to leverage the great advantages that can be gained from technology and social media. India’s state agencies, particularly the ECI and other regulatory institutions, will be put under tremendous pressure. Many analysts speculate that the 2024 elections will be the most expensive ever (Pradhan 2024). Key parties plan to spend aggressively on social media and a number of them are hiring social media influencers (Singh and Parkin 2024). It, however, remains to be seen how effectively the ECI and other state institutions will be able to monitor and check the use and misuse of social media by candidates, parties and their supporters and manage the daunting challenges of disinformation, the growing use of generative AI and deepfakes as well as hate messages, and safeguard the integrity of the electoral process.

Based on the above analysis, various considerations for policymakers, oversight agencies, social media platforms and civil society actors are set out below to enhance transparency in online campaign finance and safeguard the integrity of India’s electoral processes in the digital age.

Regulate and monitor third party spending

• Policymakers should consider introducing regulations on monitoring and limiting third party spending on digital campaigns, including expenditure by affiliated groups, supporters and interest organizations.
• The ECI should implement mechanisms to track and disclose expenditure by third parties on behalf of political entities, to mitigate the risk of undisclosed influence on electoral outcomes.

Promote compliance with the Voluntary Code of Conduct

• The ECI should design strategies to encourage political parties and candidates to adhere to voluntary codes of conduct, and promote transparency and ethical conduct in online campaigning.
• Collaboration should also be strengthened with major social media companies to enhance compliance monitoring and enforcement mechanisms, thereby ensuring adherence to agreed standards of conduct.

Enforce stricter disclosure requirements

• Existing legislation should be reviewed to mandate political parties and candidates to disclose all expenditure related to online campaigning, including payments to social media platforms, the costs associated with content development and the salaries of digital campaign personnel.
• The ECI should implement robust mechanisms at the operational level to ensure comprehensive reporting and disclosure of digital campaign expenses in the financial statements submitted to the electoral authorities.

Strengthen regulatory oversight

• Investments should be made by the ECI and other relevant bodies to enhance the oversight capabilities of responsible personnel to monitor and regulate the financing of online campaigning effectively.
• In order to promote compliance with political finance laws, sanctions should be levied on political actors for any violation, including undisclosed spending (including by third parties) and circumvention of regulations.

Improve public transparency

• The ECI could establish public databases or portals for the publication of comprehensive information on political advertisements, sources of funding and expenditure in a user-friendly interface, as well as data visualization tools to enable greater transparency and accountability.
• Transparency-focused civil society organizations could promote initiatives on social media literacy to equip citizens with the skills to critically evaluate online political content and identify misleading or deceptive information.

Lolita Cigane

13.1. Overall regulatory framework

Latvia is a parliamentary republic with legislative powers rested in the 100-seat parliament (Saeima) and executive powers—in the Cabinet of Ministers. It indirectly elects the president of the republic who mainly has ceremonial powers. The Constitution of Latvia was adopted in 1922 after Latvia declared its independence in 1918. Latvia was occupied by Nazi Germany and the Soviet Union during the World War II and remained under Soviet occupation until 1990/1991 when it declared its full independence and the legal force of the 1922 Constitution was restored.

In accordance with the Constitution, Latvia is divided into five electoral districts and has a proportional electoral system. The system allows voters to express their preference not only regarding the political party list, but also regarding individual candidates. In line with this system, each voter can express their candidate preference on a party list they have selected by crossing out the name of a candidate and thus giving the candidate one point less or putting a plus next to a candidate’s name and thus moving the candidate up the list. Only political parties can present their candidate lists; no individual candidates can stand. In such a system, intensive, personalized negative publicity can play a crucial role in a candidate’s election prospects and fosters intensive interparty as well as intraparty competition. The online environment provides for an especially fertile ground for that.

13.1.1. Evolution of campaign finance regulation

Latvia’s campaign finance system is an ever-evolving regulatory regime that responds to emerging issues and constantly identifies and resolves problems that hamper meaningful oversight.

In the early 2000s, before Latvia joined the North Atlantic Treaty Organization (NATO) and the European Union, the international community identified high-level political corruption as a core problem, which it referred to as ‘state capture’ (Hellman et al. 2000). Latvia’s civil society organizations (CSOs) began seeking ways to address the problem through systematic campaign finance data gathering, analyses and advocacy (see Open Society Justice Initiative 2005).

These efforts coincided with international pressure to create an independent anti-corruption agency, which resulted in the establishment of the Corruption Prevention and Combating Bureau (KNAB). In 2002, parliament asked CSO activists whether they would prefer the Central Election Commission or KNAB to conduct campaign finance oversight, and they advocated the latter. Box 13.1 provides a summary of the factors that have led to a strong campaign finance regulatory system in Latvia.

CSOs in Latvia conducted systemic campaign finance monitoring projects in 2001, 2002, 2005 and 2006. In the election years of 2010, 2011 and 2014, the regulatory regime was assessed by the civil society and electoral stakeholders as sufficiently developed and the oversight agency’s capacities sufficiently strengthened; so no systemic CSO monitoring was deemed necessary. Laws on campaign conduct were consolidated during this period, which facilitated the inter alia definition of third-party campaigning and introduction of bans on administrative resource abuse, hidden advertisement, and paid TV advertising 30 days before elections. Figure 13.1 depicts the general evolution of the country’s campaign finance regulatory system.

Latvia’s detailed campaign finance regulation is stipulated in two main laws—the Law on Financing of Political Organizations (Parties)²⁴ and the Pre-election Campaign Law.²⁵ Figure 13.2 summarizes the key benchmarks of the development of the campaign finance system.

In many countries, civil society and the public are astonished by the many challenges that online political campaigns present. Such campaign practices are difficult to conceptualize in laws because of overlaps with the principles of freedom of speech in user-generated content, complications in monitoring of the frequency and content of online campaign ads, and challenges in ensuring the transparency of the origin of the ad and achieving meaningful reporting of online activities by political parties, candidates and third parties. Therefore, election watchers are sometimes discouraged from talking about or pursuing the regulation of online campaign finance because of the sheer number of unknowns tied to it (e.g. OSCE–ODIHR 2022). Latvia, however, over time has passed laws that define and address the most complicated campaign issues and campaign finance is now under comparatively strict and transparent oversight. This situation has been achieved through the constant interaction between the vocal, active and articulate civil society, supportive media, responsive lawmakers and proactive oversight agency (Figure 13.3). This combination has allowed online campaigning to be regulated at a comparatively early stage and has facilitated the creation of specific rules for online campaign finance regulation.

This process, however, has been characterized by peaks and troughs of activity and capacity on all sides (Box 13.2). Opportunistic individuals and organizations have challenged the newly agreed principles of campaign finance regulations; therefore, advances in regulation and response have been followed by setbacks.

Without these consistent efforts over the past decade, Latvia’s political system would have come under serious pressure from foreign financing and influence through both traditional and online campaign means.

The law has required political parties to report all online campaign spending since 2006. Initially this referred to more traditional forms, such as advertising on Internet news portals, and later campaigns’ presence on social media. Online campaigning is now mainstreamed in the itemized and detailed political party finance reports, which provide a clear picture of the importance and growth of online campaign expenditures. Deeper analyses of these party expenditures illustrate that political parties have been experimenting with this form of campaigning; the investment in online campaigning varies over time. For instance, Vienotiba (Unity), one of the main political parties, spent almost EUR 40,000 on Internet advertising in 2011, only EUR 9,000 in 2014, and almost EUR 40,000 in 2018. Another large party, Harmony, spent approximately EUR 30,000 in 2010 and EUR 10,000 in 2014, but jumped to EUR 99,000 in 2018.

There was a considerable increase in online expenditures during the most recent parliamentary elections in 2018 (see Figure 13.4), but online advertising still constituted only 12 per cent of total campaign spending. In accordance with the assessment of the KNAB, expenditure on online campaign advertising by political actors is proportionally smaller. Moreover, political parties are disincentivized to manipulate the playing field in this way (e.g. by hiding the identity of the source, or attempting to exceed the limit), and would not encourage the participation of third parties and in-kind donors through online activity.

Table 13.1 provides a detailed overview of campaign finance regulation in Latvia and its relevance to the online campaign finance regulation. The broader purpose of this table is to prove the point that if the legal framework sufficiently addresses the campaign finance in general, it has more ability to also guide the regulatory ideas and approaches for the online environment. This table also serves as guidance for ideas of how some of the more complex campaign finance concepts could be tackled.

13.1.2. Specific online campaign finance regulation

Table 13.1 illustrates how online campaigning activity regulation is mainstreamed throughout the law on campaigning, which treats online as an equal medium along with the TV, radio, print and outdoor advertising.

However, Latvia also has had specific regulations related to online campaign activities since 2016. In the USA, for instance, despite evidence that paid as well as coordinated unauthentic online content activity from abroad influenced the 2016 election results (Priestap 2017), any mention of Internet campaigning dates back to 2006 when the practice was in its infancy (OSCE 2021: 23). Sections 18–19 of the Pre-election Campaign Law contain specific regulations pertaining to online election campaigning.

• A direct contract is required between the campaigner (political party) and their authorized persons and providers of advertising services. This prohibits intermediation, for instance, by an advertising or PR agency.
• All entities that offer advertising services online at least 150 days before election day must notify KNAB of their price lists for placing election campaign materials, including planned discounts and criteria for their application that will be used for the duration of the campaign. This information is immediately published on the KNAB website and made available to the public; prices and policies cannot be amended after this time.
• If an online advertiser does not supply a price list, they cannot place election material during the 120-day campaign period.
• When placing campaign materials online, the campaigner who paid for the ad must be clearly and unambiguously indicated.

13.2. Mandate of the oversight agency

The prime role of KNAB is to serve as a pre-trial investigative body with traditional police powers. Its mandate includes conducting criminal investigations of corruption offences (in the public sector), preventing conflicts of interest, and conducting risk assessments, awareness raising and educational activities. In addition, it also conducts political and campaign finance oversight.

Campaign finance and annual finance reports and declarations submitted by political parties are published and made available to the public through the political parties’ financing database (KNAB n.d.). KNAB has to regularly (at least once a year) inform the public of any violations of these regulations by political parties and the steps it has taken. This is an effective tool for preventing violations of campaign finance regulation. In its oversight function, KNAB can apply to the chief justice of the Supreme Court to access bank account information, to get approval for wiretapping and to obtain search warrants. It may also call and interrogate witnesses.

13.2.1. Appointment

KNAB is headed by an executive director (ED) who is nominated after an open competition and the work of the selection committee. A commission chaired by the director of the State Chancellery selects the applicants for ED. The prosecutor general, the director of the Constitution Protection Bureau and the chief of security policy are part of the commission.

The parliament appoints the ED of KNAB upon the proposal of the Cabinet of Ministers. The parliament can only appoint or reject a candidate; it cannot propose an alternative. The ED is appointed for a maximum of two consecutive 5-year terms by a simple majority vote and can be removed by a parliamentary vote upon the suggestion of an evaluation committee; the decision must be supported and passed on to parliament by the Cabinet of Ministers. The ED organizes the selection and recruitment of all KNAB employees.

All administrative decisions are taken by the ED, her or his deputies, the heads of departments of the central headquarters and territorial branches, investigators and other persons who ensure the fulfilment of the functions of KNAB and are responsible for them. The ED determines the jurisdiction of examination of matters and decisions taken by KNAB.

13.2.2. Accountability

KNAB is an independent public administration institution under the supervision of the Cabinet of Ministers. The supervision, executed by the prime minister, is limited to the lawfulness of KNAB’s decisions.

All administrative decisions taken by KNAB can be first appealed to the ED and then in the respective administrative court. Criminal investigation decisions can be appealed in accordance with the procedure set out by the respective laws applicable to all institutions that have similar mandates.

KNAB publicly reports the results of its investigations into political party financing and campaign financing at least once a year. It submits a report to the Cabinet and parliament on the fulfilment of its functions and informs the public of pre-election campaign violations and the measures undertaken to address such violations.

13.2.3. Independence guarantees

Although the prime minister supervises KNAB, this oversight is limited in scope. The prime minister (herself or himself a political party nominee) cannot revoke ‘administrative acts’, which include KNAB decisions regarding political party financing, such as a decision to impose an administrative fine on a party or order it to return illegally acquired funding. The prime minister has no authority to supervise KNAB’s criminal investigations.

13.2.4. Organization

KNAB is staffed with approximately 150 people, 10 of whom serve in the political party finance control unit. Although KNAB has operative and investigative powers, citizens and political party representatives are estimated to supply approximately 80 per cent of the evidence it evaluates regarding potential campaign finance violations, including online violations. KNAB attributes this high degree of interest and involvement to the fact that political and campaign finance issues due to intensive media and civil society attention have been in the public discourse since 2001. Therefore, the public is sensitized to the importance of campaign finance issues and their relevance to the quality of democracy (KNAB 2021). The easily downloadable and user-friendly ‘Report to KNAB’ application was developed in 2018 in the context of serious concerns about potential foreign electoral interference online. In connection with the 2018 election KNAB received 1,179 reports of potential campaign violations through this application (KNAB 2018). Another explanation for this high number is the fact that the information can be easily submitted—photo, audio or video evidence can be added without identifying the source. KNAB reports that it has sufficient capacity to process this information, and it serves as an important monitoring tool during elections. In contrast to the political culture in other former communist countries, KNAB maintains that the public as well as political parties and candidates are eager to report the suspicious campaign finance activities (KNAB 2021). This is especially important for the decentralized and fragmented online campaign world in which algorithms work to micro-target small groups and even individuals.

In sum, the fact that KNAB is a general anti-corruption agency that has campaign finance oversight as one of its functions has increased the gravitas of this role and facilitated more consequential and consistent oversight. Nevertheless, as the previous discussion attests, there are plenty of tools that an oversight institution that does not have investigative powers as part of their mandate can use to develop methodologies to address the campaign finance challenges of the online environment.

13.3. Applying general campaign finance oversight to online campaigning

After revelations of foreign influence in the US and French presidential elections in 2016 and 2017, respectively, involving the unrestricted purchase of online advertising, there were fears that a similar scenario would be repeated in Latvia, especially given its geopolitical vulnerabilities (Box 13.3). In addition, in an opinion poll conducted just before the 2018 elections, support for the statement ‘any change is better than no change’ was very high, indicating a high degree of openness to last-minute voting decisions, i.e. a high degree of voter volatility (Providus 2019).

CSOs and investigative journalists initially raised the alarm about potential foreign influence in the 2018 election and the lack of preparedness on the part of the oversight agency (Krutaine 2018). KNAB was prepared to take a narrower approach, concentrating on the practices it had developed for previous elections, even though the global examples had highlighted new vulnerabilities posed by the online environment. CSOs concluded that, ‘although with strong mandate and powers, the regulator initially was unaware that they should and could do something about campaign in social media’ (Stafecka 2019). Civil society activists proposed the following:

• Public authorities, especially KNAB, should urgently prepare guidelines for political advertising on social media, and obtain the necessary Internet tools to monitor political ads on social networks.
• Activities both on traditional media and online should be monitored.

13.3.1. Special task force

A special task force on election security coordination was created in response to civil society advocacy (Table 13.2). It was chaired by the State Chancellery—a government institution under the direct supervision of the prime minister. The task force sought to ensure election integrity and to protect the public space from undue foreign influences. The group, headed by the prime minister’s advisor on strategic communication, consisted of representatives of the secret services, KNAB and several ministries. The task force worked in several directions: it monitored the media space, established cooperation between all public authorities with a role in securing elections and elaborated action plans related to different election risks (Stafecka 2019). The task force, both on its own and together with other state institutions, cooperated with the largest social media platforms—Facebook and Google. This cooperation was needed to agree on a swift reaction in case harmful or fake content appeared. Several Facebook pages, including a fake account of a minister, were closed as a result of task force activities. Earlier cooperation had not been successful, when Latvia sent Facebook 12 information requests about 11 accounts in the first half of 2017 (Krutaine 2018).

A significant part of the task force’s work involved organizing trainings/seminars for editors representing national and regional media. Media editors are important electoral stakeholders that can block false information from being amplified and disseminated. Their awareness of this gatekeeper role constitutes an important safeguard against amplifying false information. For instance, during Germany’s 2021 Bundestag election campaign, mainstream media reported scandals involving Green Party candidate Annalena Baerbock that seriously damaged the party’s approval ratings. While the origins of the smear campaign were not fully ascertained, the German affiliate of Russia Today, known to spread official Russian propaganda, was the third-most viewed news channel online at the time (Berzina et al. 2021).

The task force served as the main cooperation body between all involved state institutions; it held strategy, response and coordination meetings and received and shared updates. A clear chain of command was established in case of emergencies. The task force’s main conclusion was that it could not prove that a foreign country had tried to interfere in a coordinated manner in the 2018 parliamentary elections.

13.3.2. The role of KNAB

KNAB bought the licences of election campaign spending monitoring tools on the largest Internet platforms. This monitoring covered paid advertisements by political parties and affiliated persons as well as other political messages (Stafecka 2019).

Cooperation with the social media platforms improved following several public diplomacy activities, including the president of Latvia visiting the Facebook HQ in Silicon Valley as part of his visit to the USA. Facebook provided all the requested information regarding campaign spending on its platform. The advertising expenses declared by the political parties and disclosed by Facebook were similar (Krutaine 2018). The successful cooperation with Facebook and Google continued in the 2019 European Parliament election (KNAB 2018).

13.4. Conclusion

Although it is tempting to attribute the success of online oversight in Latvia to KNAB’s investigative and operative powers, these parts of its mandate, although helpful, play a lesser role in its relatively successful oversight. A concise and coherent approach to systemic monitoring and constant improvements in regulatory framework have played a larger role (Table 13.3).

Although online campaigning is more difficult to capture due to its decentralized, fragmented and individualized nature and the fact that most popular social networks originate abroad, some countries have developed rigorous campaign oversight traditions and regulatory frameworks that facilitate the ability to oversee the process online as well.

CSOs and the media have played a critical role in observing, monitoring and advocating the constant update of campaign finance rules in response to the changing nature of campaign means and approaches since before Latvia joined the EU and NATO. These efforts have been met with comparatively responsive and understanding lawmakers and an oversight agency that generally implements its mandate meaningfully and consistently.

Social networks have now introduced safeguards against foreign electoral influence, and societies and regulators are more aware of the dangers such influence poses to the quality of democracy. Most widely used social networks now take precautions to sell campaign advertising only to users that originate in the country where the elections are held. In addition, there appears to be an early warning system in place to root out unauthentic coordinated behaviour of online trolls and bots. In addition, for instance during the Bundestag elections in Germany in 2021, most political parties deployed disinformation and misinformation detection and combating teams that served as an early warning system for harmful content (OSCE-ODIHR 2022).

KNAB insists that future cooperation with social media should not rely solely on gentlemen’s agreements between state institutions and social networks. The agency also seeks more guarantees that it will receive rapid replies from social networks in case of campaign law violations (Stafecka 2019).

Since the unrestrained online foreign interference in campaigns in 2016, many more safeguards have been put in place to vet the origin of user-generated content as well as paid advertising online—especially in social networks. However, in many places it remains the responsibility of social network companies. The US Congress, for instance, despite massive interference in the 2016 elections did not manage to regulate online and social network conduct before the 2020 presidential elections, leaving it to the discretion of online media outlets; some adhered to stricter internal regulation than others (OSCE-ODIHR 2021). In countries where the domestic legislation is stricter and clearer and oversight is consistent, regulators seem to be better able to obtain access to information on paid advertising (in addition to publicly available information) and remove harmful content. Latvia, as shown above, has been relatively successful at this. However, such ad hoc strategies and bilateral contracts should evolve into more systematic regional and even global responses given that an increasing number of people are using online content and the further fragmentation of the online space with algorithms that permit the highly personalized targeting of small groups of users or individuals is likely.

Octael Nieto-Vazquez

14.1. Introduction

While many political parties rely on campaign mobilization on the ground to attract voters, others employ digital strategies for electoral support. Such online tactics typically focus on targeted messages addressed to segmented audiences (Harker 2020; Nadler, Crain and Donovan 2018) that candidates and parties may see as potential supporters. This case study examines the regulatory environment in relation to political finance, including guidelines and strategies as well as norms, practices and patterns of spending on digital campaigns in Mexico. The case study also documents some recent experiences of Mexican political finance oversight agencies in dealing with the digitalization of campaigns. Finally, it reviews two emblematic judicial decisions and cases in tribunals on parties’ and candidates’ online campaign strategies. The study is based on a literature review of academic and policy papers and judicial documents as well as seven key informant interviews with advisors of the national electoral management body (EMB), campaign activists, campaign consultants and former candidates.

In recent years electoral practitioners and legislators have viewed the regulations of digital campaign spending from two perspectives—traditional and modern. The former conceptualizes online political strategies as a new branch of established models of political communication. From this standpoint, standard regulations on party expenditure must adapt to new practices, essentially by innovative ways to interpret the law (e.g. Chester and Montgomery 2017; Neudert 2020). In contrast, the innovative approach sees digital political spending as a new method for parties to attract voters, which requires a conscious reformulation of regulations. From these views, online interactions between candidates and citizens face unique challenges that risk violating political rights (Dommett and Bakir 2020; Dommett 2019, 2020a, 2020b).

Although both perspectives have contributed significantly to the introduction of spending regulations on digital campaigning, Mexican electoral guidelines provide an interesting case that conflates distinct approaches to digital political finance. In practice, the rules for campaign spending in Mexico rest to a large extent on the model of political communication used in the country, which has been shaped since 2007, and the gradual adoption of secondary statutory instruments influenced by emerging strategies for digital campaigning. Essentially, digital campaign spending in Mexico is governed by the principles of freedom of speech and free access to information and a model of political communication.

With the goals of motivating further analytical review and providing global stakeholders with insights for overseeing online campaign spending in Mexico, this case study proceeds as follows. It first explains the normative background and the current regulatory framework for digital advertisements and online campaign spending. It also examines the implications of other constitutional, financial, technological and regulatory measures for the oversight of online campaign financing. In addition, the case study analyses campaign spending from the 2018 and 2021 Mexican presidential and mid-term elections and reflects upon judicial decisions on digital party practices and behaviour. Finally, it presents closing remarks.

14.2. Background: Mexico’s normative framework around digital advertising

Online political ads in Mexico are unregulated when they are published by individuals, given that doing so is legally understood as a form of free speech. This approach recognizes digital information as free public material that voters can use when deciding how to vote. However, party spending on digital advertising is subject to regulations framed by the National Electoral Institute’s (INE) statutory instruments and the so-called model of political communication adopted constitutionally in 2007. In pursuit of fair elections, the constitutional reform introduced three strict guidelines: (a) restricting the direct hiring of traditional media companies; (b) setting the timing for the broadcasting of political advertisements; and (c) setting caps on campaign spending.

14.2.1. Freedom of speech in relation to digital campaigning

The Mexican Constitution (article 6[2]) guarantees the right to free speech by allowing every person to access, seek, receive and disseminate information and ideas by any means of expression. Based on this principle—and a Supreme Court precedent (Supreme Court of Justice of the Nation 2007)—the Federal Electoral Tribunal (TEPJF) ruled that freedom of speech applies²⁶ to the Internet and its different forms of communication. The electoral judges concluded that to guarantee the constitutional right of freedom of speech, individuals’ access to information through communication technologies, broadband and Internet services must remain unregulated. They stated that ‘universal access to broadband, broadcasting and telecommunications services allow all people access to information equally, publicly, openly, without any discrimination, thereby contributing to strengthening a society of rights and liberties based on equality’ (TEPJF 2016a: 25).

The right to free speech, however, entails the exchange and dissemination of ideas among individuals exclusively; it does not apply to political parties and candidates, as they pay to promote themselves online. The best way to understand this is by dividing ideas propagated online into two categories: those posted without interest other than sharing views and those published for obtaining votes. The electoral authorities have adopted some criteria to decide whether a digital message is subject to regulation: the costs of production involved, the timing of the message and the taxpayer status of the source of the message.

14.2.2. Implications of regulatory reforms on digital political communication

The tightening of regulations on political parties’ access to traditional media has also escalated the shift towards online campaigning in Mexico. The 2007 constitutional reform configured a model of political communication aimed at encouraging fair elections between parties and candidates. Nonetheless, this reform established a foundation for online advertising regulations and legal criteria. By amending article 41 of the Constitution, the Federal Congress responded to the 2006 post-election political turmoil motivated by ‘negative propaganda’, or ‘black campaigning’ (Buendía Hegewisch and Aspiroz Bravo 2011: 13 and 15; Damazo 2014). But the change had an enormous impact on the regulation of funds spent on campaigns and on the political advertisements that were broadcast. As a result, the reform established a new kind of interaction between political parties and the mass media—in other words, a new model of political communication.

This model also granted new powers to the electoral authorities to monitor, oversee, inspect and sanction violations; these powers gave the INE the right to investigate and punish organizations and individuals outside politics. In addition, the 2007 constitutional reform empowered the INE to enhance its oversight measures concerning money spent on campaigns. The reform has progressively shaped the way parties and candidates hire digital companies and influencers to advertise political messages.

Another aspect regulated by the 2007 reform is the content of the messages broadcast. Initially, the amendment allowed the INE to review and approve campaign messages before their being broadcast on traditional media. Based on that power, the INE developed a national programme for monitoring political messages on radio and television (INE n.d.), which has become gradually more sophisticated since 2009 in terms of reaching emerging forms of campaigning, including through social media and the Internet. Currently, parties and candidates must submit a proposal for any audiovisual promotional material for approval in advance regardless of the means of transmission. In contrast with hiring restrictions on traditional media, however, they are allowed to freely and directly pay companies and individuals for online campaigning as long as they report the transactions transparently.

14.3. Key features of Mexico’s regulatory approach to online campaign spending by political parties

To gain a comprehensive understanding of the normative framework for campaigning on the Internet and social media, we must examine INE’s statutory instruments (INE 2014). This section discusses five aspects: (a) the legal definition of ‘paid advertisement’; (b) the institutional attributes needed for the INE to develop innovative oversight measures; (c) the lawful conditions for parties to substantiate financial transactions about online activities; (d) the constitutional foundations of the non-campaign period before polling days; and (e) the legal liability of third parties regarding digital political advertisements.

14.3.1. Proving a legal definition of ‘paid digital advertisement’

The term ‘paid digital advertisement’ is defined as ‘insertions, e-banners, tweets, published messages, social media accounts, websites and other similar paid items whose purpose is to promote a campaign, a political party or a candidate’ (INE 2014: article 199[4][2]). A key element of this definition is the regulation of e-items aimed at promoting campaigns, parties or candidates; therefore, all election-related messages might be subject to scrutiny by the authorities. In addition, by law (INE 2014: article 215), parties and candidates must hold contracts and invoices for advertisements published on the Internet.

14.3.2. Empowering the INE to conduct meaningful oversight

Internet campaign items are understood as messages aimed at obtaining citizens’ support ‘in addition to those indicated in art. 76 of the Political Parties Law [such as the promotion of parties’ and candidates’ names, personas or manifestos] and those identified by the INE’s Technical Unit, based on information disseminated on the Internet, social media or any electronic means benefiting liable subjects’ (INE 2014: article 203[1]). One implication of this definition is that the INE has the right to carry out all necessary measures to validate information submitted by parties and candidates with information that it collects itself. Among these measures, the INE is authorized to request information from third parties for confirmation (see 14.3.5: Regulating third-party campaigners). Any anomaly resulting from such an action is further investigated, and sanctions may be applied if a violation has taken place. In other words, overseeing digital campaign spending in Mexico relies on the INE’s monitoring operations on the ground and the information provided by parties and candidates. Although implementing such measures tends to be expensive, interviewees agreed with their overall effectiveness.²⁷

14.3.3. Ensuring high levels of political finance transparency

Liable subjects must provide supporting documentation for their campaign expenses. In addition to reports, parties and candidates must submit valid fiscal receipts for advertising published on websites and social media with details of the ‘company or companies … contracted, dates of publication, e-addresses, costs spent, [the name of the] candidate and the campaign that benefited [and] samples of the advertisement posted on the Internet’ (INE 2014: article 246[1][e]). The purpose of this requirement is to push political parties and candidates to carry out traceable financial transactions and to avoid informal trading. This rule is also supported by other regulations on credit and financial institutions (see 14.4.2: The banking and fiduciary secrecy exception). An interesting innovation here is a data set of thousands of campaign items, a price matrix developed by the INE in which they set market values every year in order to estimate and compare the cost of items reported by parties.

14.3.4. Setting time for reflection

The Constitution (article  41[II] and [III]) states that political messages should not be broadcast during a ‘closed season’, which is a period in which campaigning is banned so that electors can reflect on their voting intentions. The TEPJF (2009) ruled by precedent that, to safeguard the principles of fair elections, the dissemination of government advertisements in the mass media is restricted in terms of content and timing. As to content, the TEPJF stated that, ‘in no case may messages be published with an electoral nature; that is, they must refrain from aiming to influence electoral preferences for or against parties or candidates’. As to timing, the precedent stipulates that ‘messages may not be disseminated in the context of elections during the periods that comprise the pre-campaign, the campaign and the period of reflection [closed season], consisting of three days before and up to the end of polling’. In the words of one interviewee, candidates are aware of both restrictions but do not necessarily follow them properly, ‘given that they usually are trying to find a way to promote themselves’. ²⁸A more comprehensive analysis of a judicial case is presented below (see 14.5.3: The judicialization of digital campaigning in Mexico) to clarify how this rule works.

14.3.5. Regulating third-party campaigners

The fiscal status of taxpayers who post campaign-related messages online plays a fundamental role in the INE’s decisions over sanctions. Since freedom of speech pertains to ideas disseminated by individuals only, when someone who makes money from online advertisements publishes political messages, that person may be accountable to the INE. In some cases, the electoral authority could assume that specific election-related messages that are not reported by parties or candidates must be considered part of campaign spending. In such a situation, the INE uses its price matrix to calculate the money spent and then unilaterally adds the estimation to the campaign that is benefiting from the messages.²⁹ Parties and candidates may face underreporting sanctions or penalties for exceeding the campaign spending cap, which could result in the withdrawal of a candidacy.

In order to trigger this procedure, the INE applies certain criteria when monitoring political messages. For example, to distinguish between a non-political from a campaign video/clip that has been broadcast, one of the elements the INE observes closely is the level of technicality and sophistication involved in the production of the item. If the video employs a drone or expensive equipment, or shows national celebrities, well-known presenters or footballers, the message is eligible for further inspection. In the case of written messages, the INE looks for campaign patterns by comparing other political messages posted by parties and candidates.³⁰ Although finding a written campaign trend may rely on a subjective interpretation, the INE has successfully sanctioned a party in a complex case about written campaign messages online, discussed below (see Box 14.1). Despite criticisms of this unilateral measure, expressed by politicians and political parties—based on hundreds of legal challenges against the INE—voters, the general public and civil society groups appear to see the action positively.³¹

14.4. Additional legal instruments for overseeing online campaigning

As explained in 14.3: Key features of Mexico’s regulatory approach to online campaign spending, Mexico recently adopted regulations for overseeing online campaigning. Nonetheless, the framework is assisted by legal instruments beyond the realm of elections which make a substantial contribution to monitoring and verifying digital advertising. Firstly, this section discusses the implications of a constitutional precept regarding the responsibility of political parties for spending state resources. Secondly, it examines a banking and fiduciary secrecy exception—a controversial measure adopted in Mexico in 2005—with considerable implications for overseeing public funds. Thirdly, it describes the development of technological tools for tracking parties’ and candidates’ expenditures, registering candidates for legal liability, registering campaign service providers, and monitoring street and printed political advertising.

14.4.1. Political parties share constitutional responsibility for monitoring digital content

Some of the legal foundations for sanctioning political parties against illicit funding rest on their constitutional definition. Given that political parties in Mexico use public money, the Constitution understands them to be ‘entities of public interest’ (article 41[I]), a definition with significant implications for judicial decisions. Political parties are organizations that are neither subject to nor subordinated to private or particular interests, which entitles the electoral authorities to sanction parties for operations with private donors and individuals who contribute to parties and campaigns.

On this basis, the TEPJF ruled that political parties must reject any contribution made by unauthorized individuals, which means that, as direct beneficiaries, political parties are fully accountable for the actions of third parties. The tribunal stated the following:

the political party is the guarantor of the conduct of its members and other persons related to its activities, in compliance with its functions and/or in the achievement of its goals and, therefore, is responsible for their conduct, regardless of the responsibility of each individual in particular, which can be internal to the organization, or exceed those limits. … This means that there can be both an individual responsibility (on the part of a natural person who is a member of the party, or of an outsider) and a responsibility of the party as a legal person in charge of the correct and adequate fulfilment of the functions and obligations of its members for non-compliance with the duty of vigilance. 
(TEPJF 2003)

In fact, this legal precept was invoked in a case against the political party Movimiento Ciudadano (Citizens’ Movement, MC), which explains the fine imposed on the party for posts published by an unaffiliated individual (see Box 14.2).

14.4.2. The banking and fiduciary secrecy exception

An enactment that helps authorities to track financial transactions is the exemption for banking and fiduciary secrecy in the Credit Institutions Law (article 142[1]). Consistent with the protection of personal data, the law regulates how credit institutions are permitted to store information about deposit holders, services and all kinds of financial operations. The instrument also establishes the basis for collecting data from depositors, debtors, account holders or beneficiaries, and representatives involved in banking operations and financial services. Furthermore, the law establishes guidelines for lawsuits or claims filed by trustees and the requirements of judicial authorities. However, the enactment stipulates secrecy exceptions by enabling banking and credit institutions to provide data and information when requested by the authorities in specific settings without a judicial order (article 142[3]).

The measure assists governmental agencies in tracking the origins and destination of funds, particularly when public money is involved. Essentially, the rule allows prosecutors or federal authorities to request information from customers of banking institutions when they face processes of review, certification or audit, for the verification of potential criminal activity, such as money laundering, fiscal fraud or other possible fiscal offences. In particular, the INE and local EMBs across the country have been expressly granted the right to request financial information from parties, candidates and individuals when auditing, inspecting and investigating anomalies in party financing (article 142[3][IX]).

14.4.3. Technology for spending oversight

In order to collect and organize financial data, the INE has developed four national systems for inspecting political parties and candidates: (a) the Audit Comprehensive System 5.0; (b) the National Registration System for Precandidates and Candidates; (c) the National Registry of Providers; and (d) the Comprehensive System for Monitoring Street and Printed Political Advertising (INE 2022).³² These platforms have assisted the INE substantially in analysing evidence to decide numerous cases involving party funds.

The Audit Comprehensive System 5.0 is an online instrument designed for liable candidates and parties to report income and expenditure operations, submit supporting documents, generate accounting reports and submit quarterly/annual reports on ordinary financial operations. The National Registration System for Precandidates and Candidates is a system in which citizens interested in running for a post submit their candidacies. This tool organizes and digitizes potential candidates’ information at the national and local levels. The National Registry of Providers is an online tool in which potential parties’ and candidates’ providers must register and submit supporting documents. The Comprehensive System for Monitoring Street and Printed Political Advertising is a tool that helps the INE to identify street displays and printed advertisements nationally and locally. Reports from this tool are subsequently compared with parties’ and candidates’ statements, and any difference is then assessed in detail.

14.5. Online campaign advertising in Mexico: How do parties and candidates report digital activities, and how do the authorities respond?

Political parties and candidates in Mexico have gradually developed more sophisticated methods for digital campaigns (Cárdenas, Ballesteros and Jara 2017: 23). Paid online advertisements during electoral cycles, however, undergo scrutiny because of existing regulations around party spending. Given the nature of the public financing system in Mexico,³³ the INE and other governmental oversight agencies are empowered to conduct inspections and hold liable individuals and organizations accountable. A remaining question is how parties and candidates report their online campaign platforms. This section examines some online campaign evidence and numbers from the 2018 presidential election and the 2021 mid-term elections in Mexico.

14.5.1. What do we know about digital campaign advertising in Mexico?

Shaped by changing forms of digital campaigning, the Mexican electoral authorities have faced challenges in developing regulatory instruments within the law. In 2014 the INE issued a set of innovative compulsory items for reporting that force parties and candidates to disclose their expenses on digital strategies (INE 2014). Firstly, the list includes no fewer than 10 entries, including ‘Internet websites’ as digital campaign activity. By 2016, however, the INE had added 26 items applicable to the 2018 Mexican presidential election (see Table 14.1). All the entries can be divided into two key dimensions: on the one hand, those aimed at identifying parties, candidates, coalitions, locations and dates, and, on the other hand, those that classify types of spending and the amounts of money spent. Initially, the INE used one item for identifying liable persons; in its latest guidelines, however, the authority established 13 entries with crucial information such as level of election (federal, state or local), electoral jurisdictions and both campaign and expense dates. To classify types of spending and amounts, the INE originally developed 10 key entries for reporting expenses on media strategies and later expanded this to 13 items (indicated by the $ symbol in Table 14.1).

Furthermore, the item for reporting online strategies changed from ‘Internet websites’ to ‘Advertisement published on the Internet’ (compare item 2 in the left-hand column with item 23 in the right-hand column); currently, this entry includes social media activities. This adjustment has had a significant impact on investigations and sanctions imposed by the INE. Since 2018 online campaigns have been evaluated based on published content—audiovisual, imagery, text or sound—rather than websites contracted. The most recent debates about online activities and legal challenges rest on this substantial change.

The essence of the mandatory expenditure entries has remained since 2014. Although most conventional means of campaigning—traditional and novel—are well covered by the set of items, some interviews confirm that it is still possible to underreport transactions or to avoid reporting them at all.³⁴ According to those accounts, this is made possible by paying in cash. Once parties and candidates manage to legally withdraw money from their bank accounts or receive donations in cash, it becomes complicated for the authorities to track the funds. One interviewee, however, also admitted that, for many reasons, the accounting operations necessary to conceal financial information had to be performed surgically to succeed.³⁵

Another possible way to underreport online transactions during campaigns is by persuading influencers who are not well known to post messages aligned with campaign strategies. Such tactics tend to be more selective and aimed at specific audiences. In some cases, the selection of influencers—interviewees suggest³⁶—is determined to a great extent by their limited knowledge of electoral regulations, particularly those associated with online campaign oversight. A judicial case discussed in 14.5.3, The judicialization of digital campaigning in Mexico, illustrates this point.

14.5.2. How much have parties and candidates spent on digital strategies?

This section examines the 2018 and 2021 Mexican elections to understand the volume of party expenditure on digital strategies. Given how problematic it is to gather information worldwide on digital party spending, the fact that the INE makes organized, classified data publicly available free of charge is commendable. Meta and Google, two of the most dominant global companies in terms of digital content, also disclose terms on Mexican election advertising. Through its Ad Library, Meta provides a transparent instrument for political advertising by offering a comprehensive searchable collection of ads published across Meta (n.d.a and n.d.b). Through Google Ads Help, the company shows how it adheres to the Mexican regulatory framework on party and governmental communication (Google n.d.).

Based on official data and reports, Table 14.2 shows the expenditure reported by candidates for the 2018 and 2021 elections. In aggregated terms, we can see that for the 2018 elections, presidential candidates spent nearly a quarter of the entire campaign cost on digital strategies, more than USD 10 million. In contrast, congressional candidates disbursed a proportion of less than 10 per cent on online campaigning, slightly more than USD 5 million. An interesting aspect of the numbers is the fair parity between the items ‘street displays’ and ‘Internet’, given that the former represents one of the oldest and most traditional methods of campaigning while the latter is the most contemporary approach for persuading voters. In both elections, the disbursement share is somewhat similar; for the 2018 presidential elections, candidates employed 27 per cent versus 23 per cent, whereas congressional candidates reported 7 per cent versus 8 per cent of their funds. Another interesting point relates to the comparison between ‘canvassing’ and ‘Internet’. During the presidential contest, candidates reported less than 9 per cent versus 23 per cent of the total, whereas congressional candidates expended nearly double, 18 per cent versus 8 per cent. This difference suggests diversified methods of electioneering according to the levels of elections—federal, state and local.

While aggregate numbers indicate collective campaign trends, individual reports suggest more strategic campaign methods of candidacies. Table 14.3 shows that both proportions and amounts spent by each presidential candidate on Internet strategies during the 2018 election are staggeringly dissimilar. Whereas José Antonio Meade Kuribreña reported over USD 5 million in online spending, 24 per cent of his total campaign budget, Andrés Manuel López Obrador—the current president—declared spending USD 274,682.75 on Internet strategies, 3 per cent of his total expenditure. Only one presidential candidate spent more than half of their overall spending reported for the 2018 Mexican election on Internet strategies. Although the impact of digital campaigning on voting behaviour is not discussed in this case study, these findings provide some contrasting insights for the debate on the power of manipulating voters through social media campaigning (Kruschinski and Bene 2022; Dommett and Bakir 2020; Harker 2020; Neudert 2020; Yablon 2020; Dommett 2019, 2020a, 2020b; Nadler, Crain and Donovan 2018; Chester and Montgomery 2017; Goodman et al. 2017; Williamson, Miller and Fallon 2010; Langston and Benton 2009). Yet, as pointed out, for former candidates interviewed here, it is possible to underreport digital campaign expenses.³⁷

14.5.3. The judicialization of digital campaigning in Mexico

As described earlier, spending on digital advertising is regulated by specific guidelines. In order to illustrate how these guidelines operate in practice, this section examines two emblematic cases decided by the TEPJF on digital media framed by the 2021 Mexican mid-term elections: (a) against the national campaign of the Green Party (PVEM) (INE 2021a); and (b) against the state campaign of the Citizens’ Movement party (Movimiento Ciudadano, MC) (INE 2021b). Both incidents were reported to the INE, which imposed fines after conducting exhaustive investigations. However, the parties appealed the fines to the TEPJF, which ratified the sanctions in the first case but overturned them in the second one.

The tribunal’s decision to overturn the sanctions against the MC and the candidate was controversial because it was possible that the election results would have to have been nullified otherwise. Two interviewees said that the judges decided under pressure given that a gubernatorial candidate—the winner and current incumbent—was involved.³⁸ Also, by the time the INE determined its sanctions, the candidate had already won the election by a wide margin. As a result, reversing the overall outcome on the grounds of social media posts was arguably complicated. Although the judges admitted misbehaviour on the part of the party, the candidate and his wife, ratifying the INE’s decision might have made it necessary to call a new gubernatorial election. Most of the people interviewed for this case study agree that the TEPJF faces political pressure in cases like this, as once an election is concluded with ample margins, it is difficult to uphold claims of wrongdoing that would reverse the election results.³⁹

In the case of the MC, the interviewees said that the messages the candidate’s wife published on social media could have shaped public opinion to the benefit of her husband’s campaign. At the same time, they said that the official investigation and sanctioning of a gubernatorial candidate are notoriously slow compared with the pace at which information is disseminated on the Internet. In addition, they said that the intention to punish an elected governor who will eventually acquire political immunity puts the electoral authorities in a difficult position. Although the INE has the power to issue injunctions against candidates and parties during campaigns and pre-campaign periods, avoiding impunity depends on the possibility of withdrawing candidacies before election day.

14.6. Final remarks

The Mexican case provides some lessons on institutional design, the regulatory system and party strategies for digital campaign advertising. Based on the discussion in this case study, a few crucial aspects are listed below.

1. Internet content and social media posts are highly complex areas for the authorities to monitor. The first lesson is there is a need to develop transformative, robust institutional measures for overseeing novel methods of campaigning and party spending, such as digital strategies.
2. A second lesson is that the effectiveness of institutional regulatory frameworks is highly dependent on mechanisms on the ground for challenging the information provided by parties. Consistent with the negative perception of parties worldwide—in particular, when politicians manage money—there is a need to devise sophisticated rules beyond the realm of elections to force candidates to disclose private information and make all their transactions transparent.
3. Another lesson addresses the ethical responsibility of non-political groups when involved in digital campaign activism. Since parties and candidates might be tempted to employ influencers to promote campaigns, successful regulation is highly dependent on individuals’ social reputations.
4. When parties violated the reflection period (closed season) by posting digital content, they compromised the principles of fair elections in Mexico. Although the INE implemented swift measures to ensure that the posts were taken down, these violations serve as a warning that digital campaigning might turn into a threat to elections in specific settings. EMBs might lack the powers and the budget needed to investigate and impose criminal sanctions, given their administrative nature.
5. Another aspect is the complexity of sanctioning candidates—who could eventually claim political immunity—for digital content. Messages can spread online in a fraction of the time needed to investigate and prove that an offence has been committed. Successful measures should address the withdrawal of candidacies, for example, rather than trying to remove elected officials. Once elected candidates swear their oath of office, it is nearly impossible for the authorities to hold them accountable.

Victor Adetula

15.1. Introduction

Political finance systems worldwide have had to adapt to new realities and developments at various levels, but experiences have varied from country to country. In some cases, changes have sought to address scandal, corruption or betrayal of public trust; in others the escalating cost of politics or a desire by governments, especially in newly established democracies, ‘to nurture free elections’ (Alexander 2003: 5). To varying degrees, most countries have also introduced information and communications technologies (ICTs) into their electoral systems, which has implications and consequences for the process, practice and outcomes of elections, including political finance.

In Nigeria, sections 47, 50(2) and 62(2) of the Electoral Act 2022 give the Independent National Electoral Commission (INEC) legal backing to introduce technological upgrades. Thus, in the context of the Act, INEC has been able to develop and deploy new technologies that have automated certain aspects of the electoral process and procedures (see INEC 2024). As critical stakeholders, political parties and election candidates are also adjusting to the new realities of digitalization. They are using social media platforms, notably Twitter (now X), Facebook and Instagram, to promote themselves and their campaign, as well as engaging with digital media companies to design campaign materials, using Internet platforms for fund-raising activities beyond national frontiers, and receiving donations and contributions through financial transactions driven by digital technologies.

The deployment of digital technology in Nigeria’s electoral processes has drawn attention to various loopholes in the country’s campaigning and political finance regulations. For example, the prevailing weak or non-existent regulatory framework on the use of social media makes it easy to spread disinformation and hate speech during elections. During the 2023 general elections, for instance, a letter claiming that an investigation had been launched into the All Progressives Congress (APC) presidential candidate, Senator Bola Tinubu, in relation to drug-related accusations, purportedly written and signed by an INEC spokesperson, was made public through social media. However, the letter was a forgery that originated on WhatsApp and was circulated rapidly across social media platforms. Unregulated access to Internet platforms is a major concern in the country.

There is overwhelming evidence in Nigeria’s electoral history that unregulated political finance threatens electoral integrity. This case study examines the regulatory environment in relation to political finance, from guidelines and strategies to norms, practices and patterns of spending on digital campaigns in Nigeria. It also documents some of Nigeria’s recent experiences with the deployment of digital media campaigns and assesses the performance of the country’s political finance oversight agencies. Finally, it examines the future of digital campaigns, the use of financial technology (fintech) in financial transactions during elections, and strategies for regulating the adoption of digital technologies. The study is based on a literature review of academic and policy papers, as well as 12 key informant interviews with selected officials at INEC, political party officials, CSO representatives and activists.

15.2. Political and legal framework

15.2.1. General political and electoral context

Nigeria is a federation of 36 states and the Federal Capital Territory (FCT). It has a population of approximately 230 million. The 1999 Constitution of the Federal Republic of Nigeria, introduced by the military that governed the country for almost half of its existence since 1960, retains the federal system handed down by the British colonial administration at independence. However, the Constitution introduced a presidential system with a bicameral legislature.

The evolution and development of the Nigerian electoral system have closely followed the country’s constitutional development path, dating from the colonial period when the elective principle was first introduced into the legislative assembly. The country’s electoral system passed through various phases before reaching its present state, based on universal adult suffrage, an open ballot and the first-past-the-post electoral system. The electoral system reflects the values of the most recent reforms, which include party primaries, inclusive electoral processes, the use of electronic technology for voting and the transmission of results, and accountability and transparency on campaign finances. INEC is the main electoral management body. In February 2023, it organized the seventh general elections since the country returned to civilian rule in 1999, in which 18 of the 19 registered political parties participated. The 2023 general elections were conducted under the 2022 Electoral Act, which introduced new elements into the country’s campaign finance regulations.

15.2.2. Main political finance and related laws

The structure of political finance and its regulation in Nigeria have changed markedly since the pre-independence period. The laws regulating party and campaign finance are guided by the 1999 Constitution of the Federal Republic of Nigeria (as amended), the 2022 Electoral Act and INEC’s 2022 Regulations and Guidelines for Political Parties. In addition, there are concurrent statutory provisions such as the 2000 Corrupt Practices and Other Related Offences Act, the 2004 Economic and Financial Crimes Commission Establishment Act, the 2006 Advance Fee Fraud and Other Fraud Related Offences Act, the 2020 Money Laundering (Prevention and Prohibition) Act, the 2020 Companies and Allied Matters Act (CAMA) and the 2022 Terrorism (Prevention and Prohibition) Act. The plethora of laws, regulations and guidelines regulating how electoral politics is financed in the country reflect the concerns of critical stakeholders regarding the need for electoral integrity in the country.

15.2.3. Main sources of party income and funding

In the past, political parties in Nigeria were funded by INEC using public funds. However, public funding is no longer available to political parties. Political parties receive funds from membership fees, nomination fees, selling the party manifesto and souvenirs, donations and individual contributions. The funds raised from party membership fees, or the sale of the party manifesto and other materials, are minuscule and parties must therefore seek alternative sources of funding. Political parties are not allowed to hold any funds outside of Nigeria or to retain funds sent to them from abroad.

Private donations

There is contradictory information with regard to permissible sources of donations to political parties and candidates. Chapter VII, Part I, section 221 of the 1999 Constitution states that: ‘No association, other than a political party, shall canvass for votes for any candidate at any election or contribute to the funds of any political party or to the election expenses of any candidate at an election’. This means that only political parties can fund their campaigns, solicit votes for their candidates and incur expenses for their campaigns. Any other association, such as a non-profit organization, a religious group or a business, is prohibited from engaging in such activities. At the same time, section 88(8) of the 2022 Electoral Act states that: ‘No individual or other entity shall donate to a candidate more than N50,000,000’, which would appear to allow entities to donate up to NGN 50 million (USD 35,145) to support their preferred candidate. Similarly, section 90 of the Electoral Act 2022 contains several provisions that allow for donations from individuals and entities. It is, therefore, unclear whether corporations or other legal entities can contribute to political parties or candidates. In addition, while political parties cannot receive funding from foreign sources, there is no such prohibition for candidates.

Self-funding and nomination fees

Most individual candidates are responsible for funding their campaigns, covering all campaign expenses. This is a crucial source of operational revenue for political parties, by which they fund their activities, including organizing rallies, events and outreach programmes. For instance, two leading political parties, the APC and the People’s Democratic Party (PDP), accrue huge revenues from nomination fees from prospective candidates. During the 2015 general elections, the PDP’s presidential election nomination fee was NGN 22 million (USD 110,792), while the APC’s was NGN 27 million (USD 135,972). In 2019, the nomination fee for the presidential election was NGN 45 million (USD 147,183) for the APC and NGN 12 million (USD 39,250) for the PDP. In 2023, the APC set the presidential election nomination fee at NGN 100 million (USD 242,600), governorships at NGN 50 million (USD 121,300), Senate nominations at NGN 20 million (USD 48,520), House of Representatives nominations at NGN 10 million (USD 24,260) and State Assembly nominations at NGN 2 million (USD 4,852) (Ojo 2022). The PDP imposed similar fees: presidential NGN 40 million (USD 97,040), governorship NGN 21 million (USD 50,945), Senate NGN 3.5 million (USD 8,490), House of Representatives NGN 2.5 million (USD 6,065), and House of Assembly NGN 600,000 (USD 1,455) (Ojo 2022). The nomination fees for those seeking political office are considered excessive and out of reach of those with limited financial means, especially young people and women (Olukotun 2022; Itodo 2022). The commercialization and monetization of party primaries, representation and participation in electoral processes arguably restrict these to moneyed elites that wield influence in the political sphere because of the power of their purse (Olukotun 2022).

15.2.4. Spending limit

The 2022 Electoral Act sets expenditure limits for candidates and political parties during an election. The campaign expenditure limit for presidential candidates is NGN 5 billion (USD 3.5 million), and the campaign expenditure limit for political parties is NGN 4 billion (USD 2.8 million). For candidates for the National Assembly, the campaign expenditure limit is NGN 70 million (USD 49,203) for senatorial candidates and NGN 40 million (USD 28,116) for House of Representatives candidates. The maximum campaign expenditure limit for candidates for governor is NGN 1 billion (USD 702,907). The regulations and guidelines on election expenses, particularly campaign expenses such as advertisements, campaign materials and campaign rallies, are quite elaborate but do not adequately cover the spending of political parties between elections. While there are spending limits for candidates and political parties during elections, there are no such limits outside the electoral period, which makes election-related spending limits redundant. Reports indicate that political advertisements started appearing on social media long before the official campaign period for the 2023 presidential election (Alayande 2022).

15.2.5. Regulation of third-party campaign financing

Third-party funding of campaigns is permitted in Nigeria but must conform to specific disclosure requirements. Third parties must report any campaign expenses to INEC and disclose their funding sources. There have been several instances of abuse by groups and individuals using third-party funding to circumvent regulations on spending limits. In preparation for the 2023 general elections, even before the party primaries, many groups and collectives sprang up with the sole aim of boosting support for their preferred candidates. These support groups claimed to have funded nomination forms and even bought cars as part of their support (Ojo 2022). The challenge of tracking financial transactions made in cash by third parties has further complicated effective monitoring. Until recently, Nigeria was essentially a cash economy. Efforts by the Central Bank of Nigeria (CBN) to implement a nationwide cashless society have met with resistance from Nigerians, who have concerns about electronic payment platforms. The policy was suspended in 2017 and was recently being reintroduced in stages.

15.2.6. Reporting, disclosure and oversight

INEC is the main oversight body in Nigeria, and it has the power to monitor political finance and investigate related violations. It has the statutory responsibility to audit the accounts of political parties, monitor compliance with donation and expense limits, and report any political finance infractions. According to INEC guidelines, political parties must publish their audited returns in at least two national newspapers within six months of the election, and these audited returns should also be made available to the public by INEC. The Constitution empowers INEC to deregister any political party that violates section 225, which includes failure to submit financial reports as prescribed in the Constitution and the Electoral Act.

15.3. Main trends in political and campaign finance in Nigeria

Electoral spending in Nigeria has been increasing year on year. During the 2015 general elections, INEC estimated that political parties in Nigeria spent nearly NGN 196 billion (USD 982 million) on contesting various offices, almost twice the amount spent in 2011 (Daily Trust 2018). During the 2019 presidential election, the two main parties—the APC and the PDP—accrued an NGN 11.6 billion (USD 38 million) bill in campaign expenses (INEC 2021). INEC provides further insight into the spending on traditional media by the two major political parties during the 2019 federal elections: the PDP/Atiku coalition spent approximately NGN 2.9 billion (USD 9.5 million), while APC/Buhari spent NGN 4.6 billion (USD 15 million). These figures exclude other expenses, such as transporting party cadres to rallies, the cost of hiring campaign venues, and obviously the cost of illicit electoral activities such as vote buying, bribery, hiring thugs to intimidate opponents, stealing ballot boxes and disruption of the election generally (INEC 2021: vxl).

15.3.1. Political advertising on social media platforms

Digitalization and the ‘digital effect’ are rapidly changing Nigeria’s political and social landscape. In January 2023, Nigeria boasted nearly 122.5 million Internet users, with an Internet penetration of 55.4 per cent. The number of active social media users increased from 27 million in 2019 to 31.6 million at the start of 2023 (Kemp 2023). WhatsApp is the most popular social media platform, and it is used by 91.9 per cent of social media users in Nigeria (Kemp 2022). Facebook is the most preferred platform for accessing news online, which probably explains why President Goodluck Jonathan of the PDP declared his intention to run for president on Facebook on 15 September 2010. Other presidential candidates in the 2011 general elections, notably Muhammadu Buhari of the Congress for Progressive Change (CPC), Nuhu Ribadu of the Action Congress of Nigeria (ACN) and Alhaji Ibrahim Shekarau of the All Nigeria Peoples Party (ANPP), had a strong presence on Facebook and other social media platforms.

Recent developments in fintech digital payments also suggest notable progress. The CBN national domestic card payment scheme, AfriGo, launched in partnership with the Nigeria Inter-Bank Settlement System (NIBSS), now operates in the same space as Mastercard, Visa and Verve by Interswitch. Digital payments in Nigeria are ‘expected to show an annual growth rate of 13.70 per cent’ (compound annual growth rate 2023–2027) (Emejo 2023), resulting in a projected total transaction value of USD 27.89 billion by 2027.

With this level of digital penetration, recent elections in Nigeria have witnessed remarkable growth in the use of digital and social media platforms for political and election campaigns. However, determining how much political parties and candidates spend on online campaigns presents significant challenges. Several factors contribute to these challenges. There is a low level of compliance with spending limits and disclosure requirements by political parties and candidates and only limited itemized detail in relation to reporting and disclosure requirements. In addition, the data, digital and media companies hired to carry out campaigns often work anonymously. The Labour Party is a case in point. It has admitted extensive use of social media platforms for campaigning, especially among the party’s youthful supporters. However, it claims that the party has no budget for such campaigns and that the remarkable upsurge in the use of Internet platforms for campaigning and canvassing was spontaneous rather than an organized strategy. Like similar claims by other political parties when questioned about their budgets for digital media campaigns, this claim cannot be easily verified or disproved.⁴⁰

15.3.2. Use of digital service providers for campaigning

Rumours about election-influencing services in Nigeria have been circulating since 1999. However, there has been no conclusive evidence of their use by candidates or political parties until recently. It is now common knowledge that political parties and candidates engage with data analysis service providers to collect citizens’ personal data and information in order to design personally targeted campaigns to influence their voting behaviour.

It is important to note that the deployment of digital technologies for political purposes, or the hiring of digital service providers, is not done in the same way by all political parties and candidates. In the period leading up to the 2013 governorship by-election in Anambra, for instance, the APC worked with AKPD Message and Media, a public relations consulting group that uses quantitative research techniques, data modulation and party branding to influence voters’ decisions. The party Chair later confessed that he did not understand the value AKPD could add to APC campaigning efforts until it demonstrated its expertise in that by-election (Akande 2020: 480). AKPD assessed the party’s political fortunes in the by-election and predicted disaster. The APC candidate lost, just as AKPD predicted. The company subsequently became a critical component of the APC’s thinking and planning machine as it prepared for the 2015 general elections.

The APC maximized the benefits of digital media in its 2015 presidential campaign and was ahead of others across all Internet platforms, followed closely by the PDP. Other parties contesting the presidential election used the Internet only sparingly. In the absence of ideological divides among the political parties, the struggle for political power in Nigeria is characteristically focused primarily on personalities, ethnic groups and geopolitical zones. Thus, electoral politics is carried out through political mobilization, manoeuvring and stage management (see LeVan 2019). The AKPD’s approach, however, was different. It advised the APC to stay away from a focus on sentiment and to ‘keep talking about change and jobs’, stressing that ‘the largest reason voters would support APC at that time was that they wanted change and APC should continue to say ‘“APC means change” over and over again’. Based on the polling conducted by the AKPD, the ‘voters were telling them that they wanted to hear more of “change for jobs”’ (Akande 2020: 481). The AKPD also advised the APC on which geopolitical zones to invest in for the 2015 elections, that ‘General Buhari must be at the top of the presidential ticket’, and ‘that APC should stay away from personal attacks on Jonathan and focus its critique of him on “not doing his job well”, particularly by not creating the jobs Nigeria needed’ (Akande 2020: 483). The PDP feared losing the 2015 presidential election. In a panic, it cried foul and accused the APC of using the services of a ‘propaganda machine’.

Nonetheless, as the 2015 general elections drew closer, the PDP needed to enhance its presence in the digital media space, and the political consultancy firm Cambridge Analytica came to the rescue. Cambridge Analytica is widely reported to have meddled in the country’s elections, using covert methods and the dark arts of disinformation to distort reality, as well as confidential material stolen from the campaign of APC presidential candidate Muhammadu Buhari to discredit him and boost the electoral fortunes of the PDP and Jonathan (see Kaiser 2019).

There are many unanswered questions on the funding sources for the campaigns of almost all the political parties. While there is no concrete evidence linking the procurement of external services by either the APC or the PDP to illicit funds, rumours have circulated about the use of state treasuries by serving governors and senior government officials to finance political parties during the 2015 elections. In addition, widely reported incidents of financial misappropriation, such as the alleged looting of funds from the CBN under the Jonathan administration during the 2015 elections, including a purported withdrawal of USD 2.1 billion, raise concerns regarding the legitimacy of the funds used by the PDP to engage services such as those of Cambridge Analytica (Akande 2020: 484). The use of these firms continues, including in the most recent elections in 2023.

15.3.3. Disinformation and misinformation

While digital technology platforms have made a significant contribution to the political discourse, there have been instances where they have been used directly or indirectly to attack opponents, spread false rumours and hate, and incite violence, which have had serious consequences. The volume of misinformation circulated online during elections in Nigeria, particularly by paid social media influencers, could have misled many into unfounded political judgments. The tactics used by social media influencers to shape the political behaviour of prospective voters include promoting political agendas through hashtag manipulation, deliberate mistranslations, impersonation, and manipulating audio and video material. For example, while the 2011 general elections were said to be among the fairest in Nigeria’s history and credited with remarkable use of social media platforms, they were also among the country’s bloodiest. It is believed that the use of social media to attack opponents and spread false rumours and hate contributed to the violence, as well as tensions witnessed before, during and after the elections in many parts of the country (Okoro and Adibe Nwafor 2013).

The waves of misinformation during national elections create lucrative opportunities, particularly for young people. The youth in Nigeria have become increasingly socially conscious and politically active, using social networks and contemporary information technology, including the Internet, which has promoted the proliferation of electronic information exchange. Young people find opportunities on every popular social media platform to promote their preferred candidates for elections, engage in informal campaigns and unofficial debates, and sometimes engage in misinformation, disinformation, fake news, hate speech and character assassination. Some of the most prominent political influencers on social media are hired by political parties and candidates and can earn up to NGN 500,000 (USD 1,090) a month. Those with smaller followings are likely to receive near NGN 50,000 (USD 109). Some influencers also receive lavish gifts, government contracts and political appointments for spreading false narratives and stories to enhance the chances of their clients. None of this spending is ever reported by the political parties or candidates. According to a BBC report, an influencer can receive up to NGN 20 million (USD 44,000) for delivering a result (Nwonwu, Tukur and Oyedepo 2023).

15.4. Loopholes and implementation gaps

Seven general elections and a sizeable number of by-elections and off-cycle elections have been held in Nigeria since the return to electoral democracy in 1999. Over the years, many improvements have been made to the country’s electoral system alongside great strides regarding the use of ICT tools for election management. However, threats and risks associated with the use of technology and political and campaign finance persist.

15.4.1. Limits of the legal and institutional framework

The legal and institutional frameworks for election management and administration in Nigeria lack sufficient regulation over campaign and political party finance, thereby failing to effectively curb the influence of wealthy elites over the country’s politics. There are numerous loopholes and contradictions in the legislation, particularly regarding political donations from legal entities as outlined above, which create confusion and the potential for abuse. Moreover, the existing regulations are silent on the role and impact of digitalization on political and campaign finance, and the legislation that governs public communication and laws against fake news and falsehood are barely enforced. The National Broadcasting Commission (NBC) has powers to regulate and control the broadcast industry, and the Nigerian Communications Commission (NCC) regulates the communications industry according to the Nigerian Communications Act (Cap. N97, Laws of the Federation of Nigeria, 2004). The 2000 Corrupt Practices and Other Related Offences Act, the 2004 Economic and Financial Crimes Commission Establishment Act, the 2006 Advance Fee Fraud and Other Fraud Related Offences Act, the 2020 Money Laundering (Prevention and Prohibition) Act and the 2020 Companies and Allied Matters Act (CAMA) also play a role. While these laws address some dimensions and elements of digital campaigning and political finance, they are not well focused on helping to control and regulate digital political finance and digital media in Nigeria.

15.4.2. Challenges and risks of the ‘free’ digital space

As highlighted above, despite its increased use for campaigning purposes, social media remains an unregulated space. While the Constitution bans anyone other than a political party from campaigning for political parties and candidates, in practice anyone can place political advertisements on social media in support of or opposing a political party or candidate without complying with the rules governing political finance. The costs involved in third parties placing advertisements are often undisclosed. While it is possible to argue that everybody has an equal chance in an era of digital technology, the size of the purse determines the scope of a party’s or candidate’s opportunities and the benefits they gain. In that regard, political parties and candidates with resources are able to spend more both offline and online, procure consulting groups for digital media services and conduct a more influential campaign with a broader reach. There is a saying in Nigeria, particularly among members of the political class, that ‘elections are not won on social media’. However, using different social media platforms to share views on candidates cannot be without benefit to those sponsoring online campaigns.

15.4.3. Inadequate capacity for oversight enforcement

The INEC, through its Election and Party Monitoring Department (EPM) and under the supervision and guardianship of the Election and Party Monitoring Committee (EPMC), has developed a mechanism for tracking, collating and reporting campaign expenditures by parties and candidates. While INEC’s efforts to track campaign finances during the 2015 general elections were unsuccessful, these improved for the 2019 general elections. It inaugurated an Inter-agency Campaign Finance Monitoring Group (IACFMG) to promote synergy between INEC and other relevant agencies in monitoring the sources of funding and spending of all candidates and political parties to ensure they did not exceed the approved ceiling.⁴¹ The group comprised the Economic and Financial Crimes Commission (EFCC), the Financial Intelligence Unit, the CBN, the NBC, the NCC, the Centre for Social Justice, the Nigeria Police Force, the Federal Inland Revenue Service, the Independent Corrupt Practices and Other Related Offences Commission and the Nigerian Press Council. Its inauguration was well received, but the IACFMG became a talking shop that rapidly fizzled out as it did not receive adequate support from INEC and other relevant agencies.⁴² For example, some member agencies had no budget to undertake the tasks that fell within their mandate.

For the 2023 election, INEC deployed observers to monitor the campaigns of political parties and candidates to ensure that they complied with all laws, regulations and codes of election conduct. It was meticulous at tracking expenditure on billboards, posters and jingles in the media to estimate funds spent by both the parties and the candidates. In addition, INEC provided campaign finance tracking forms to political parties, candidates and media houses for them to complete and submit within six months of the general elections (INEC 2024: 122). Despite these efforts, INEC faced several challenges in relation to effective oversight. Among these were problems with its technical workforce, a lack of proper coordination with other relevant agencies, and the limited or absence of information from political parties and candidates. At no time was it able to mobilize and deploy an adequate number of personnel to monitor and track political parties’ income and expenses effectively.⁴³

Moreover, enforcement of the financial regulations on political parties was weak, partly due to the lack of clarity in some of the provisions of the political finance laws and regulations. Apart from a few cases in the 1950s and 1960s, the laws on campaign finance have not been tested in the courts. No political parties or candidates have ever been sanctioned for political finance infringements. Greater clarity is also required on the enforcement regime. For example, it is not clear whether INEC is authorized to impose fines after administrative hearings or, if it can, how this power fits with the powers of the courts.

15.4.4. Lack of will of political actors

There is also the problem of a lack of political will on the part of political actors, often expressed in their laid-back attitude. Political parties and candidates have been found to have violated political finance regulations with impunity and have exhibited a total refusal to complete and return their expenditure tracking forms. Several months after the 2023 general elections were concluded, for example, no financial report on the expenditure of political parties in compliance with the campaign and political finance laws and regulations had been submitted to INEC by any political party. When they were submitted, several parties failed to provide adequate information on the expenditure incurred by candidates, making it impossible for INEC to properly monitor party and candidate campaign finances (INEC 2024: 228).

At the same time, the National Assembly has not demonstrated sufficient support for effective control and regulation of campaign finance. For example, article 226 of the Constitution establishes that INEC must investigate a political party’s accounts and report to the National Assembly if it considers that the accounts have not been maintained properly. Article 228 outlines the powers of the National Assembly to impose sanctions. However, no sanctions have ever been imposed for the many political finance infractions identified by INEC (Adenekan 2022). Article 228 also states that the National Assembly may provide an annual grant to INEC to be disbursed to political parties equitably to help them carry out their functions. However, the National Assembly has never provided a grant to INEC to enable it to reduce the excessive dependence of political parties on private funding from moneyed interests.

15.4.5. Low-level capacity of civil society

Civil society groups are critical stakeholders in the Nigerian electoral process. However, Nigeria’s experience of engaging with CSOs and coalitions during electoral processes is very recent. It was only in 1999 that civil society groups got their first taste of organized electoral participation, and the scope of their involvement has since been expanded. In terms of monitoring political party and campaign finance, however, their engagement is still marginal and uncoordinated. They lack the organizational capacity and necessary resources to monitor campaign finance and effectively engage with government, political parties and INEC to advocate for better transparency and accountability and the necessary reforms. Moreover, political parties have become increasingly intolerant of civil society, especially in campaign finance, and are unwilling to share their financial reports with CSOs. Some efforts have been made by international organizations to strengthen the capacity of civil society and media outlets to demand transparency and accountability in campaign finance, but more needs to be done.

15.5. Next steps and recommendations

It is possible to argue with some conviction that the prospects for effective regulation of campaign and political party finance in the digital age are bright. The commitment of INEC’s leadership first to electoral reform and second to the cyber sphere represents a landmark for Nigeria. There is also increased consciousness across the country, most noticeable in official circles, of the benefits of digitalizing critical processes and procedures in the political arena, as expressed in the spirit of the pillars and objectives of the National Digital Economy Policy and Strategy, 2020–2030 (NDEPS). This aims to leverage digital technologies for economic growth and to lift millions of Nigerians out of poverty (Federal Ministry of Communications and Digital Economy 2020). The activities and operation of the Ministry of Communications and Digital Economy, which has been mandated to develop and implement harmonized and coordinated policy on the digital economy for Nigeria, align with the above vision (Ibietan 2023: 3–4). Similarly, the digitizing of critical activities and operations in the country’s public services, such as the use of Bank Verification Numbers, Treasury Single Accounts and the Integrated Payroll and Personnel Information System (IPPIS), signals considerable promise. In addition, the proactive positioning of the CBN as the country’s apex bank for addressing the loopholes in digital currency transactions through checking, monitoring and regulating to prevent misuse, fraud and illicit transactions is highly commendable.

On the other hand, the country’s domestic context provides a significant constraint and some dominant social forces in the country demonstrate a lack of interest in genuine political finance reform. Among the major challenges that face reform of political finance in general and online campaign finance in particular are: (a) the growing level of political corruption linked to the dominance of a money-driven political culture that pervades almost every area of social life, supported and encouraged by a continuing attraction to ostentatious living by many; (b) a lack of political will and of any readiness on the part of the political class to undertake and support reform; and (c) a general apathy, and lack of capacity and courage, on the part of civil society in the country to engage in a reform project.

Strengthen regulatory frameworks

In view of the weakness of existing regulatory frameworks, efforts need to be made to close the gaps, such as addressing the challenges posed by digital technologies and monitoring campaign finance. The virtual space in Nigeria is almost a ‘free for all zone’ and the lack of an effective regulatory framework compounds the situation. This has implications for the scope and intensity of the penetration of digital technologies into the electoral process. Any future electoral reform must be broad enough to factor in the realities of the digital era and include regulation of social media platforms, the procurement of digital media services and the hiring of influencers, among other things. In addition, the future legal and institutional framework should underscore the importance of effective coordination and collaboration between different agencies beyond merely setting up joint committees and specify the modality of cooperation, resource allocation and capacity development.

Strengthen the capacity of INEC

Strengthening the financial and human capacity of INEC will be crucial to ensuring effective monitoring of political finance and enforcement of penalties and sanctions. For instance, enforcing the spending limits provided for in the regulations will require information on election spending to be systematically collected and collated for analysis. Various aspects of political finance oversight, including online campaign finance oversight, require specialist knowledge and competencies such as on database management, data analysis, social media communications, cybersecurity and information security, the digital economy and fintech. A proper audit of existing capacities must be undertaken to understand the training needs of the officials responsible for political finance monitoring and oversight, and subsequent programmes should be introduced to meet those needs.

Increase constructive collaboration among the oversight institutions

While INEC is the primary agency responsible for monitoring campaign finance, effective enforcement of the political finance regulations requires effective collaboration and synergies with other agencies with unique capacities and resources in specialized areas and functions. For example, the involvement of the EFCC in the screening of candidates for election should be encouraged, and the capacity of the Code of Conduct Bureau needs to be strengthened and it should be involved in monitoring the assets of public servants. In the same way, the CBN, the Nigeria Data Protection Bureau, the National Information Technology Development Agency, and other relevant agencies involved in monitoring and recording financial transactions need to develop a culture of working in collaboration with INEC. The effectiveness of inter-agency coordination must be enhanced by providing a clear mandate for each agency. Furthermore, the collaborating agencies need to be provided with adequate resources to function effectively, which includes equipping them with the capacities required to monitor and track online campaign finance.

Strengthen the capacity of political parties

The processes and outcomes of the 2023 elections demonstrated rapidly increasing deployment of digital media services by political parties and candidates. However, experience of and adjustment to the digital space varied from one party to another. For example, while new political parties such as the Labour Party emerged as successful players in the digital environment and could adapt their processes from the outset, established parties, notably the APC and the PDP, had not fully adjusted to the digital environment. Future electoral reform must address how political parties adjust to the realities of digitalization. This could make the task of monitoring, controlling and regulating the operations of political parties less cumbersome for INEC, which may need to insist on adopting various applications, programs and tools for ensuring transparency. While essential legislative reforms are being discussed, political parties themselves can take initiatives to improve their processes, practices and actions. For example, it is a tradition among the political parties to come together as the general elections draw closer to jointly develop a code of ethics, usually on how to conduct themselves and avert a breakdown of law and order. This understanding could be extended to adopting best practices on the use of digital and social media for campaigning.

Strengthen the capacity of civil society organizations

The CSOs in Nigeria are increasingly acknowledged as watchdogs, maintaining vigilance over the electoral process. Their monitoring ensures that campaign finance practices are transparent, and they are also actively engaged in advocacy for political finance reform. Very few CSOs are currently engaged in these activities as they lack the capacities. This warrants the design of systematic capacity-building programmes, particularly in the areas of tracking and monitoring online campaign finance, analysing the financial reports of political actors and running advocacy campaigns, as well as civil education programmes on the negative consequences of irregular party financing, among other things.

James Venslauskas

16.1. Introduction

Technological advancements are changing the landscape of political campaign finance in the United States—now shaped by cryptocurrency (digital currency), social media advertising, social media influencers and artificial intelligence (AI). The popularity of social media platforms enables campaigns in the USA to take a digital-centric approach to sharing and promoting political content, and to increasingly adopt non-traditional methods of raising funds for political purposes. The Federal Election Commission (FEC) plays an essential function as the autonomous regulatory body tasked with administering and enforcing the country’s campaign finance law at the national level (FEC n.d.b). This case study will address opportunities and challenges at this scale, without delving into a detailed discussion of state-level dynamics. However, it is important to acknowledge that states independently govern campaign finance and political advertising within their jurisdictions. Some state-level examples will be referenced pertaining to the effectiveness of the regulatory framework and to closing gaps and loopholes in the age of digital political finance.

The objective of this case study is twofold: first, to document and analyse strategies, practices and challenges pertaining to the digital aspects of US political finance and campaigning. Second, to document the lessons learned, to assist stakeholders in safeguarding the integrity of institutions and political processes in other contexts. While the USA has a relatively lax regulatory framework for digital political finance, knowledge gathering in this field is a pressing concern for voters concerned about electoral accountability and the risks to democracy, including digital risks, corruption and foreign interference. The case study begins by providing an overview of the FEC’s legal framework on political communication. Subsequently, it explores social media advertising—these ads can be misleading, so knowing who is funding these ads, what their interest is, and how much they are spending helps voters make informed decisions during elections (Lund and Strine 2022; Campaign Legal Center n.d.) and the role of digital service providers and digital currency in political campaigns. Following this exploration, an analysis of the effectiveness of the regulatory framework is presented, along with a concise section on the lessons learned and concluding remarks.

16.2. Legal framework

The USA is a federal republic forming a presidential system with three branches of government: the executive (the President), the legislative (Congress), and the judicial (US Supreme Court). The state level has a similar structure of government. A pivotal moment in federal politics occurred in 1971 with the enactment of the Federal Election Campaign Act by the US Congress. A principal legal framework governing political donations, the legislation regulates political spending and imposes limits on campaign fundraising. Subsequently, the establishment of the FEC in 1974 introduced disclosure mandates for campaign contributions and granted corporations, labour unions, and trade associations the ability to form Political Action Committees (PACs) (Caltech n.d.). PACs receive support from their members to facilitate campaign activities, subject to federal constraints on funds raised and spent. However, political finance for state senators and representatives is governed under state laws and regulations. These state regulations vary widely, especially concerning digital ads and platforms.

While the FEC is primarily responsible for regulating federal campaign finance, other institutions, such as the US Supreme Court, exert significant influence. The landmark 2010 case Citizens United v. Federal Election Commission demonstrated the complicated nature of US campaign finance laws. The Supreme Court ruled that corporations could contribute unlimited amounts to ‘electioneering communications’ (media distributed publicly that references a federal candidate around the election season) (FEC n.d.c). Additionally, the US Court of Appeals for the DC Circuit created the legality of what are known as ‘Super PACs’ in SpeechNOW.org v. FEC (2010) (Persily, Bauer and Ginsberg 2018). Super PACs are corporations and groups outside the 1971 framework described above. They can freely accept and spend unlimited sums of money on political ads, often lacking sufficient transparency regarding the origins of these donations (Lau 2019). In the decade after its entrance into campaign finance in 2010, spending through Super PACs has increased from USD 62 million to USD 2.1 billion in 2020 (OpenSecrets n.d.b; Persily, Bauer and Ginsberg 2018). As of 1 November 2024, Super PAC spending in the 2024 electoral cycle was USD 2.6 million (Open Secrets 2024).

On 1 December 2022, the FEC adopted a rule on disclaimer requirements on Internet communications, which broadened prior requirements and definitions. Previously, Internet requirements for political ads and communications were loosely regulated with a lack of concrete regulations on disclosure requirements (FEC 2022a). Conventional political advertising media, particularly television, already had clear regulations on disclaimers. These must be clearly written (and sometimes audible) and understandable for viewers to recognize who paid for the advertisement and if it was authorized by the candidate’s committee. There are specific requirements for how long the disclaimer must be displayed, how much space it should take up, and even the contrast between the text and background (FEC 2022b).

The updated disclaimer requirements for Internet communications aim to align with those applicable to broadcast and print media, tailored to the unique dynamics of online communication. Under the new rules, which took effect on 1 March 2023, the FEC notes that ‘Disclaimers must be “clear and conspicuous’’ regardless of the medium in which the communication is transmitted.’... ‘identifying who paid for the communication …’ (FEC n.d.d). The FEC clarified the definition of public communication to cover ‘communications placed or promoted for a fee on another person’s website, digital device, application or advertising platform’ (FEC n.d.d). One major difference from other media is that public communication on the Internet is not required to have the ‘stand-by-your-ad’ disclaimer, which requires political candidates to personally associate themselves with their television and radio ads to discourage controversial claims or attack ads. Instead, online ads must have a written disclaimer that ‘can be viewed without taking any action’ if the form of communication has a text or graphic element. However, the new rule has exceptions for Internet public communication that would allow for adapted disclaimers or exceptions to disclaimers altogether (FEC n.d.d). These new requirements are more comprehensive than previous rules, but loopholes persist for digital ads, as the FEC’s limited enforcement capabilities can raise concerns about transparency and accountability.

Introduced by Senator Amy Klobuchar (Democrat-Minnesota) but not yet passed by either chamber of Congress, the ‘Honest Ads Act’ of 2023 (see GovInfo n.d.) goes further than current FEC rules by proposing requirements for disclosure on digital advertisements, a ban on online platform advertisements paid for by foreign nationals, and an official database of online political ads.

16.3. The rise of social media advertising in political campaigns

The rapid evolution of political advertising technology—coupled with weak regulations of disclosure—reduce electoral transparency and can contribute to erode public trust. Currently, there are no regulations in the USA requiring purchasers of online advertising to disclose their expenditures and donors, marking a lack of transparency (Weiser and Bannon 2018). Additionally, estimating expenditure on political advertising is challenging due to inconsistent disclosure requirements and varying policies among social media platforms regarding what constitutes a political advertisement. These gaps in transparency can make ads misleading; understanding who funds them, their interests, and their spending helps voters make informed decisions during elections (Lund and Strine 2022; Campaign Legal Center n.d.).

Digital platforms have transformed campaign political communication and advertisement as a result of changing news and entertainment habits. Fundamental to this transformation is the use of social media and digital services such as streaming services. Social media platforms advance communication, especially bridging gaps between voters and politicians. Social media brings benefits to democracy and civic engagement, but it also poses risks, especially in campaign finance. A major issue is the lack of transparency and enforcement in political advertising, allowing corporations to use ads to influence electoral outcomes in ways that serve their special interests.

The importance of social media in campaign finance is shown by the spending of the leading presidential candidates in 2020, Donald J. Trump and Joseph R. Biden. In total, some USD 2.1 billion was spent on online political advertisements (since 1 January 2020 on Facebook; and since 27 May 2018 on Google). Altogether, 14 million ads were produced with the involvement of 98,000 advertisers (OpenSecrets 2024). As noted in Table 16.1, the Trump Make America Great Again Committee from January 2020 until January 2021 spent USD 94.6 million on Facebook advertisements and USD 46.2 million on Google (OpenSecrets n.d.e).

During the 2020 campaign season, Biden’s team spent USD 108.2 million on Facebook advertisements and USD 83.7 million on Google, totalling USD 191.9 million on online advertising (Table 16.2) (OpenSecrets n.d.c). The Trump campaign spent approximately USD 129.7 million on Google and USD 138.7 million on Facebook, totalling USD 268.4 million (OpenSecrets n.d.d) during the 2020 campaign season (Table 16.3). Google and Facebook will likely become even more important as fewer voters use broadcast and print media (Massoglia and Evers-Hillstrom 2019; Schwartz 2022). An election expert informant for this case study reiterated the idea that digital platforms are garnering significant small donor contributions (Palmer 2023).

The US voters are overwhelmingly in support of limits on campaign contributions by individuals and groups, with 72 per cent in support, according to Pew Research Center (Cerda and Daniller 2023). What is striking is the ease with which individuals and groups can contribute to campaigns and public communication through digital means. For example, the Super PAC named Americans for Prosperity Action received a USD 3 million donation from the multinational conglomerate Koch Industries in August 2022 and went on to spend over USD 810,000 on Facebook advertisements (Schwartz 2022). These substantial donations and the use of social media platforms for political advertising can damage public confidence in government and threaten democratic principles. As concerns among US voters rise over the influence of money in politics, the limited regulation and transparency in political advertising spending can contribute to a perception of distorted representation. This lack of oversight may deepen public skepticism and erode confidence in the fairness of electoral processes.

16.4. Digital strategies in political finance

According to industry group the Interactive Advertising Bureau (IAB 2021), a social media influencer is defined as an individual or group of content producers who have amassed a social media following due to their ‘personality, lifestyle, content style, writing and/or opinions’ (TealHQ n.d.). Political parties and candidates strategically recruit social media influencers for their ability to amplify political messages to a target demographic. In 2023, 90 per cent of Americans were using social media, the leaders among the adult population being Facebook (74.2 per cent), Instagram (60.7 per cent), TikTok (42.4 per cent), and X, formerly known as Twitter (41.8 per cent) (Shewale 2024). Distinct from traditional print and broadcast media, digital platforms allow for influencers’ political messaging to appear immediately and in a manner more relatable to their particular followers. Besides celebrities, small-scale influencers—also referred to as nano-influencers (fewer than 10,000 followers)—are ‘normal individuals … who have connections to their followers offline’ (Goodwin, Joseff and Woolley 2020: 5). Small-scale influencers are thus able to convey a sense of authenticity and trust in their social networks, extending this to their favoured candidates.

Ethical considerations nevertheless arise surrounding financial transparency and integrity of influencers’ political messaging. Importantly, payment transactions between political campaigns and influencers often remain secretive as social media platforms are unable to govern off-platform dealings. In particular, PACs can further increase their political contribution efforts by hiring social media influencers through ‘dark money’ that remains undetected by governing authorities (Caltech n.d.). The term ‘dark money’ refers to spending aimed at influencing political outcomes where the origin of the funds remains undisclosed (OpenSecrets n.d.a). In 2020, more than USD 116 million of ‘dark money’ was linked back to Democratic or Republican party-affiliated groups; existing FEC loopholes allow for such contributions to remain concealed from the public eye (Massoglia 2020). The current FEC framework fails to enforce disclosure of campaign influencer payments on social media platforms. Although Super PACs are required to disclose their donations, certain groups come to function as de facto ‘dark money’ outlets when the majority of their funding cannot be traced back to the original donor (OpenSecrets n.d.a). As well as being a corruption risk in itself, this lack of transparency creates a challenge to the level playing field: some political campaigns disproportionately benefit, undermining the principles of fairness and integrity in the democratic process. It also increases the risk of foreign interference in elections. The lack of transparency can not only enable corruption but also disrupt the fairness of political campaigns. Concealed funding can create misleading impressions and promote biased messaging through influencers. This can undermine the electoral process and can compromise democratic principles. When campaign contributions remain hidden, voters struggle to identify the sources of political influence, impairing their ability to make well-informed decisions. Consequently, this lack of clarity can distort public perception and can contribute to weaken the overall credibility of elections.

Having previously banned political advertisements due to controversies, X announced that it would lift the ban in preparation for the 2024 US presidential election. X’s Chief Executive Elon Musk reported the platform would expand its safety and elections teams, deploy screening processes for eligible campaigns, and establish an advertisement transparency centre for anyone to review promoted political content—without seeking to ‘determine the truthfulness of disputed information’ (Woollacott 2023). While X will impose stricter policies on paid-for-promoted political posts, campaign-hired influencer content on the platform will remain difficult to unearth. Moreover, Elon Musk’s public endorsement of one of the 2024 US presidential candidates raises concerns about potential conflicts of interest and the platform’s ability to maintain impartiality. Other social media companies emphasize fact-checking to contain the spread of election misinformation, but may be outpaced in this by evolving influencer marketing strategies. Stories that are falsified or debunked have often already made their impact; by the time they are corrected, viral content and mobilized follower support have generated widespread attention and significantly influenced public perception, making it challenging to fully mitigate the damage. As mentioned, no federal guidelines exist concerning influencer political advertisement so it is unsurprising that social media platforms’ disclosure policies lack detail (Lai 2022).

The NextGen PAC, the largest youth voting organization in the USA, funded around USD 2 million during the 2022 campaign cycle in support of Democratic candidates. It partnered with TikTok influencers who were required to disclose their paid posts (Lai 2022). The Democratic Majority Action PAC paid influencers between USD 300 and USD 500 per post, while other campaigns opted to hire influencers to join their staff (Lai 2022). Not all PACs and public relations firms require influencers to acknowledge paid political content in their posts. Determining paid versus unpaid influencer content mainly rests on a trust system that allows for deceptive messaging and questionable ethical practices. When influencers act as paid PR, they may promote messages they do not genuinely believe in, misleading voters in the process. This deception undermines electoral accountability, as voters may not realize they are being influenced by paid endorsements rather than genuine opinions (Zakrzewski 2024). Such practices allow campaigns to propagate false messages at arm’s length, eroding their responsibility and potentially compromising the integrity of the electoral process.

The emergence of cryptocurrencies has further transformed political finance in the USA. Decentralized currencies such as Bitcoin bypass traditional banking systems, breaching spending limits. Similar to the regulatory uncertainty of paid social media influencers, cryptocurrencies also lead to concerns of illicit contributions, anonymity and political financial accountability. The decentralized structure of cryptocurrencies enables peer-to-peer transactions to be validated without an intermediary. Additionally, cryptocurrencies offer enhanced security through cryptography and user anonymity, which further complicates tracking fund activity.

The cryptocurrency industry has greatly increased its lobbying of Congress in recent years, spending USD 21.6 million in 2022 compared to USD 2.5 million in 2020 (Giorno and Sayki 2023). In 2022, Senators Debbie Stabenow (Democrat-Michigan) and John Boozman (Republican-Arkansas) introduced the Digital Commodities Consumer Protection Act (DCCPA) which proposes a supervisory administrative body for ‘digital commodities’, as well as to oversee the cryptocurrency industry’s Commodity Futures Trading Commission (CFTC) (Chen 2022). The DCCPA would strengthen federal authority as it overrides state-level registration prerequisites regarding money transfers and virtual currency. The collapse of the cryptocurrency exchange platform FTX, and the indictment of its founder, Sam Bankman-Fried, underscores the severity of securities fraud and money laundering. Notably, Bankman-Fried was a significant supporter of the DCCPA, which highlights the ongoing tensions and regulatory challenges within the cryptocurrency industry (Chen 2022; Giorno and Sayki 2023). Pertinent to the FTX case, customers’ funds were illegally used to finance political campaigns (Department of Justice 2022). This underscores a significant issue in contemporary political finance, emphasizing the potential adverse impact of cryptocurrencies on the US political landscape. Furthermore, cryptocurrencies make it easier for cross-border and global financial donations to political campaigns, and raise the risk of foreign interference. Digital anonymity conflicts with democratic governance, which relies on open and accountable processes.

16.5. Analysis of regulatory effectiveness

Regulatory frameworks on digitalization in political finance and campaigns exist in the USA, however, many believe they are inadequate in addressing transparency and accountability in the digital age (Zakrzewski 2024). The FEC finalized a ruling on online political advertisements only recently—while they have been in use for a decade—indicating the slow pace of policy development at the national level. While the FEC’s new regulations on disclaimers for online political ads are paving the way for transparency at a future point, loopholes and vague definitions remain. Transparency and accountability are the primary concerns of political finance. Many experts working in this field agree that the priority in the digital age is how to strengthen and encourage transparency.

An election expert interviewed for this study commented on the ambiguous delimitation of public communication, which makes it difficult to identify what a political ad is and what the FEC is responsible for regulating. Similarly, the Harvard Law Review draws attention to the FEC’s referring only to political ads as those ‘placed’ for a fee, whereas previous drafts had mentioned ads ‘placed or promoted’ for a fee (Harvard Law Review 2023). Furthermore, disclaimers for ads on ‘services’ were also rejected in the final regulation covering public communications. It is unclear whether these rejected phrases will create areas of unregulated communications. More generally, the same Harvard Law Review report indicates scepticism about enforceability and effectiveness of these rules.

An expert in campaign finance voiced similar concerns about gaps in the regulatory framework that could easily be exploited by Internet-based platforms and applications. Ports (2022) shares the concern about stable interpretation and enforcement, pointing out the same changes in wording—and that FEC Commissioners will issue ‘Interpretative Statements’.

The FEC enforcement process can be in the form of audits, complaints, referrals or self-submissions (FEC n.d.a). However, the party-political composition of this body greatly hinders these processes, especially at a time when bipartisan collaboration is at a low ebb. No more than three commissioners can belong to the same party, yet four votes are required for official FEC action (Weiner and Bacskai 2023). These partisan divides leave the FEC with weak enforcement capability.

Another of the expert informants emphasized free speech in the political process, including digital ads and political contributions. Regulation may infringe on free speech or human rights. To counter the risks of AI or other technology, they recommended educating the public on how to identify misleading information and political speech.

Perspectives from experts interviewed for this study thus differed greatly. The views of anti-corruption advocates tend to emphasize the bias of the system towards wealthy individuals. They point out that restricting individual contributions is impractical and emphasize how money has not only dominated but also corrupted politics since Super PACs were created (Vogl 2023).

Because nationwide regulation on political finance and campaigns on digital platforms has been inadequate in addressing major concerns, some in the field of campaigns and political finance suggest state regulatory frameworks may be more adept at enabling transparency (National Conference of State Legislatures 2023; Weiser and Bannon 2018). Some states such as California, New York and Washington already have laws that require social media platforms or online media to disclose information about contributors to political ads and to record information on a public database (National Conference of State Legislatures 2023). If a number of states adopt their own policies, it may create ripple effects across the country. However, it is uncertain whether this decentralized approach will achieve the desired outcomes. At present, regulatory and legislative efforts on this issue are insufficient and lack comprehensive scope.

Social media companies, too, have deployed tools to balance the transparency and privacy of their users. Facebook does not fact-check political ads but has instituted an Ad Library which gives information on how political ads are targeted and how Facebook political campaigns are financed (Hutchinson 2020). Users can withdraw from audience targeting, stop categorization based on collected data and opt out of political ads—if not entirely (Hutchinson 2020). Facebook’s advocacy for greater regulation has included support for the aforementioned ‘Honest Ads Act’ proposals (Hananel 2023; Hutchinson 2020).

Facebook’s development of transparency tools further highlights the shortcomings of the 2023 disclosure rules. Unlike the usual approach that starts with the actor (campaign, political party, PAC), addressing non-transparent ads produced by the Russian Internet Research Agency required Facebook to shift the focus to the ad itself, and to determine the political nature of the content by taking a broader approach to ad classification, signalling a proactive self-regulation approach on the company’s part (Harbath 2023).

Besides Facebook, Google is the other major platform in the political advertisement arena as other social media companies including TikTok, LinkedIn and Pinterest have banned campaign advertisements (Nott 2020). Google’s policies for political advertisements are consistent with those issued to other products and it is against their guidelines to issue false messaging (Nott 2020). Furthermore, Google’s election ads are designed to align with traditional media practices, making them more visible and accessible for public discussion (Nott 2020).

However, the key commercial players in digital political advertising space are not legally liable for the content posted on their platforms. This is because under section 230 of the Communications Decency Act 1996, social media companies are considered Internet service providers and not publishers (Nott 2020). Unlike newspapers and television stations, which are protected by the First Amendment’s guarantee of freedom of the press, social media platforms are private companies. Therefore, they are not obligated to adhere to these constitutional protections and can set their own content policies⁴⁴ (Nott 2020). Facebook prominently displays ‘Paid for by’ labels on political advertisements to disclose funding sources, hence the availability of campaign spending data such as that in Tables 16.1–16.3 above. Records show two primary US presidential contenders in 2020, President Joe Biden and Former President Donald Trump, spent a combined historic USD 217 million on advertisements across Facebook and Instagram (a subsidiary of Facebook) (Stromer-Galley et al. n.d.).

An expert in social media election integrity emphasizes the intricate challenge of discerning the authentic sponsors behind Facebook political ads. The 2016 election led to significant criticism of inadequate collaboration between the government and social media companies. Addressing this issue necessitates the implementation of robust guardrails, especially given the complex nature of managing domestic speech, which encompasses the expression of ideas, opinions and information by individuals and entities within the country’s borders (Harbath 2023). Ideally, congressional-level legislation could provide a unified regulatory framework across states, ensuring transparency for all stakeholders.

The potential consequences of unregulated AI could play out in the 2024 general election. While many aspects of AI could be used in campaigning, ‘deepfakes’ are currently gaining the attention of politicians and the FEC. In August 2023, the FEC held public comment on proposals to ban deepfakes in campaign advertising. Deepfakes are ‘computer-created manipulation of a person’s voice or likeness using machine learning to create content that appears real’ (McKenzie 2023). The obvious danger AI and deepfakes pose to informed voter choice is the capacity to deceive audiences that candidates or other political actors have spoken or acted in various ways detrimental to electoral credibility. In June 2023, Republican presidential candidate Ron DeSantis released a manipulated video featuring Former President Donald Trump, which was an obvious ‘attack ad’. Most concerning, however, are deepfakes that are not readily perceived as such, for at least some portions of the electorate. Deepfakes are also a potential tool that could be used by foreign entities to interfere in US elections, as noted by the Congressional Research Service (Sayler and Harris 2023). While the FEC is hearing comments on deepfakes, it is evident that AI is already being used in the 2024 campaigns with little regulation (McKenzie 2023). AI will be a significant concern for potential threats to campaigns and political finance as it develops and becomes more convincing.

Sarah Kreps, specialist in AI policy’ reports that ‘erosion of trust is particularly dangerous in a political context. Nowhere is this issue more pressing than with the 2024 election cycle’ (Cornell University 2023). Google announced that from November 2023, it would require political campaigns to label ads that utilize AI (Cornell University 2023). The decision to require advertisers to include disclaimers—such as ‘This audio was computer-generated’ or ‘This image does not depict real events’—will support responsible political content dissemination on Google’s platforms (Alba 2023) and may set a constructive example for other social media companies to emulate.

President Biden’s Executive Order (EO) on the Safe, Secure, and Trustworthy Development and Use of AI, unveiled in October 2023, is a step forward in seeking to foster responsible innovation within AI technology (The White House 2023) (see Box 16.1). However, it is characterized by numerous voluntary commitments; several provisions lack the force of law. Additionally, the EO neglects the domain of political finance, leaving unresolved questions about tackling fraudulent financial activities that may persist or emerge through AI. With federal regulatory bodies failing to act quickly to provide guidance on how to handle emerging software technologies that can mislead millions of online users and voters, it is in the hands of social media companies to implement security policies capable of distinguishing inauthentic content. The absence of stringent regulation renders these efforts ineffective. This regulatory weakness highlights a broader issue: the inability of current frameworks to adapt swiftly to new challenges, leaving significant gaps in oversight and accountability.

Given the slow pace of federal legislative and regulatory action, it is unlikely that much new regulation will be able to combat current and future risks to campaigns and political finance in the digital space, in time for the 2024 race. Existing self-regulation by social media companies will have to suffice for now. Moving forward, a more holistic approach will be necessary to combat these risks and promote innovation and opportunities in the digital space. Voters across both major parties and various demographics believe wealthy individuals have too much influence on politicians, while people in their districts have too little influence. The current lack of transparency in campaign finance may be contributing to the record low levels of trust in the federal government (Jones 2018; Pew Research Center 2023). The majority of voters support campaign finance laws and believe it is possible to limit the role of money in politics (Cerda and Daniller 2023). Thus, while immediate regulatory action is limited, there is strong public support for future reforms to enhance transparency and accountability in political finance.

One informant called for a complete ban on cryptocurrency exchanges due to links with criminal activity but another was cautious, calling for bans and regulations, citing the existing laws in place that address these concerns (Palmer 2023; Vogl 2023). Both seem to highlight the need for a robust civil society but in different ways. The anti-corruption specialists call for a civil society ombudsman to monitor and reveal disinformation on social media platforms where news proliferates at a much larger scale. Some election experts offer the opinion that the public should be trusted to understand what is real and what is not, rather than having any regulations on these platforms, limiting their freedom of speech (Palmer 2023; Vogl 2023).

Both perspectives offered insights about possible next steps that agencies, organizations and civil society could take. One perspective emphasizes not only educating the public to seek trusted sources but also ensuring fair oversight and transparency in the use of AI (Palmer 2023). Anti-corruption advocates echoed election officials' views on the need to educate the public about the dangers of money in politics and misleading news. One specialist argued that social media companies should adhere to ethical standards similar to those for print media, pointing to Russian interference via social media in the 2016 election as an example. (Vogl 2023).

Regulations and policies alone will not prevent unethical and illegal practices in political finance and campaigns in digital spaces. However, an integrated approach involving a variety of stakeholders can lead to meaningful and effective policies that ensure transparency and enforcement. Social media and digital platforms have already started implementing their own policies on political ads, while some politicians are taking pledges not to accept any donations from big PACs. Despite these efforts, voluntary self-regulation has not been sufficient to solve the problem entirely. Therefore, combining these efforts with robust regulations and collaborative enforcement mechanisms is essential to address the complexities of political finance in the digital age.

16.6. Conclusion

Political finance in the age of digitalization is a pressing matter across the USA as cryptocurrencies, social media platforms and generative software developments enable political campaigns to bypass many traditional media regulations and transparency measures. Importantly, federal oversight agencies are lagging in their efforts to institute policies in a timely manner to safeguard digital users from inauthentic or misleading political content. Amid the poorly regulated digital space, it is often the case that digital platforms are first to act and establish guidelines for political ad targeting and messaging. Without disclaimer information referencing paid or manipulated political content, identifying genuine digital information will progressively become more challenging. While the role of the FEC is centred on finance laws of political campaigns, its policies must encompass the digital sphere with updated language to govern monetary transactions that circulate outside traditional banking institutions. Worryingly, the lack of transparency inherent in cryptocurrencies and difficulty in enforcing compliance measures can open the way to foreign or otherwise illicit contributions interfering with the US election process.

The growing presence of social media influencers marks a new trend in how political campaigns in the USA are conducted. Influencers’ ability to relay information and opinion in a persuasive manner attracts a dedicated online following, and thus has the potential to serve public participation in politics (Zakrzewski 2024). However, failure to disclose paid political content can mislead followers, one among several ethical concerns. Social media platforms balance the interests of varying users, yet without transparency tools, political messaging can spread misinformation and allow disproportionate digital presence to wealthy political campaigns. Without urgent regulation of AI in the sphere of political campaigning, deepfakes threaten to create public confusion. It is a collective effort of federal oversight agencies, social media platforms, AI companies and other engaged stakeholders to unearth illegal digital practices and ensure transparent and accountable political financial transactions.