This paper is part of 'A Manchurian candidate or just a dark horse? Towards the next generation of political micro-targeting research’, a Special issue of the Internet Policy Review.

Introduction

As political campaigns compete, they try to outsmart each other by all sorts of actions: from dropping witty puns during a televised debate, to strategically knocking on doors and convincing voters. Technological innovation can help political parties improve the effectiveness of their campaigns. By using technology to collect, process, and analyse information about voters, campaigns can improve their knowledge about the electorate. Subsequently, technology can extend campaigns' capabilities of targeting specific groups with tailored messages resulting in more efficient campaigning. We call this phenomenon 'political behavioural targeting' (PBT).

Several scholars have researched political behavioural targeting in the US context (e.g., Kreiss, 2012, 2016; Nielsen, 2012; Hersh, 2015). However, the US differs in several obvious ways from most European countries. One can imagine that differences in electoral systems, privacy laws, and party financing influence campaigns’ ability to collect, process, and use personal voter data. Therefore, the findings from these studies do not necessarily apply to European countries. As there is little research in a European context, it remains unclear to what extent and how campaigns in a multiparty democracy, such as the Netherlands, use PBT-techniques. Also, it is unclear if and why there are differences between parties. In line with Colin Bennett (2016, p. 261), we wonder: "can political parties campaign in Europe as they do in North America?"

Such a question is relevant, as some scholars fear that the use of data and targeting techniques hinders public deliberation (Gorton, 2016), weakens the mandate of elected officials (Barocas, 2012), has negative effects on citizens’ privacy (Howard, 2006; Rubinstein, 2014; Tene, 2011), and enables campaigns to send tailored messages directly to citizens, thereby avoiding scrutiny from journalists (Jamieson, 2013). As a result, campaigns can potentially make opposite promises to different people, without anyone noticing.

This article sheds light on how Dutch political campaigns adopt and use PBT-techniques. Through interviews with campaign leaders, using a grounded theory approach, we answer the following overarching research question: What barriers and facilitators for the adoption and use of PBT-techniques do Dutch political parties perceive?

Theoretical framework

We will first summarise innovations in political campaigns over time, leading up to the advent of political behavioural targeting. Then, we identify the factors influencing the adoption of PBT on a campaign team level. Finally, we explore the factors that can shape the adoption of PBT on the level of national systems.

Innovations in political campaigns

Political campaigns have continuously been adapting to technological developments. Pippa Norris (2000) describes how the advent of television and the shift from partisan newspapers to national television news triggered a process of modernisation in the way political campaigns operated. Notable consequences of this shift were the adoption of a media-centred strategy in order to set the agenda, the rise of political marketing, the collection and use of data (such as opinion polls) to "shape, fine-tune and monitor campaign efforts" (Blumler, Kavanagh, and Nossiter, 1996; p. 53). Another shift came with the internet and the new possibilities for party-voter interaction that came along with the medium, which led campaigns to a new stage of the modernisation process: the postmodern campaign (Norris, 2000).

It would be an oversimplification to point to 'the internet' as a game-changer in political communication, because of the rapidly changing nature of the internet itself. As David Karpf (2012, p. 640) notes: "the internet of 2002 has important differences from the internet of 2005, or 2009, or 2012". Accordingly, much more than the advent of the internet itself, it is the advent of social media such as Facebook (2004), YouTube (2005) and Twitter (2006) which provided political campaigns with new ways of communication with the electorate (e.g., Gibson & McAllister, 2011; Conway, Kenski, & Wang, 2015; Vaccari, 2012). Together with companies such as Google, whose core business is actually not its well-known search engine but rather its advertisement business, social media not only facilitate new ways of communication, but also the tracking and collection of behavioural data of internet users (Zuiderveen Borgesius, 2016). This technique ("behavioural targeting") originates from the advertisement business. Ad agencies monitor people's online behaviour and combine this information with consumer data provided by data brokers, to target them individually with tailored ads (Turow, 2011, p.75). When applying this concept to the political realm, we can dub this phenomenon as political behavioural targeting (PBT).

Of course, PBT is not about selling products but about winning votes. And political campaigns have different means to do so than advertisement agencies have (e.g. canvassing efforts); which means that PBT happens offline as well as online. We distinguish traditional canvassing from PBT-canvassing if campaigns are able to process information about individual conversations (such as the voter's likelihood to vote for a party or her most important voting consideration), and subsequently use that information to gain strategic insights about the electorate and/or to target the voter at a later stage with a tailored message, while skipping the 'wrong' doors in a neighbourhood (Kreiss, 2016; Nielsen, 2012).

Arguably, the use of PBT can be seen as the latest step within the modernisation of political campaigns. However, as we have seen in earlier phases, not all parties in all countries adopt new techniques at the same pace and rate. Below, we identify the factors influencing the adoption of PBT. We organise these factors at two levels: (1) the individual campaign around a candidate/party and (2) the national system (i.e., the electoral system, regulatory framework, and culture). This translates into the model shown in Figure 1, which will be elaborated on in the next paragraphs.

Figure 1: Factors influencing the adoption of PBT

The campaign team level

In his extensive research of US political campaigns, Daniel Kreiss (2016) identified four factors concerned with technological innovation within political campaigns. There are resource factors, such as campaign budgets and the number of volunteers a campaign can employ; infrastructural factors, such as technological tools or skills within the organization; organizational factors, such as organisational culture and structure; structural electoral cycle factors, such as election results. Building upon Kreiss' factors, we add an additional four (one campaign team level factor and three system level factors) to examine the use of PBT. On a campaign team level, the factor is ethical and legal concerns, such as normative reservations towards PBT. On a system level, the factors are electoral context, regulatory framework,and culture (discussed below). These new factors were identified through a review of literature about innovation in data-driven political campaigning techniques (e.g., Anstead, 2017; Kreiss, 2016; Jungherr, 2016; Hersh, 2015; Nielsen, 2012), and literature about (hybridisation of) campaign evolvement (e.g., Lijphart, 2012; Plasser & Plasser, 2002; Karlsen, 2010; Norris, 2000).

Resource factors

The main elements within this factor that could influence the extent to which campaigns can use PBT-techniques are: the budget and the effort needed to carry out a PBT-operation. A large budget enables campaigns to hire skilled personnel, acquire data, or buy targeted ads. The same dynamic applies to the number of volunteers a campaign can mobilise: having a lot of them facilitates a campaign in collecting data by canvassing, and sending potential voters targeted messages (the use of volunteers, of course, is dependent on their skills). Having a small budget and few volunteers, consequently, can be a barrier for campaigns because it bars them from acquiring the same amount of capabilities or from carrying out an operation on a large scale. This is in line with normalization theory (Margolis & Resnick, 2000), according to which the possibilities of the internet will not upset traditional power structures, but will rather develop along traditional lines as in the 'offline world'.

We can also view PBT as a means of using a campaign's resources as efficient as possible, to ensure parties do not spend money and effort on voters who will vote for another party anyway, or on citizens who will not vote altogether. Then, parties with limited resources could be more inclined to use PBT to not waste precious money, time, and labour. This is in accordance with the idea of equalization, which views the internet as an empowering tool for smaller parties due to its low costs and its new ways of direct communication with the electorate (Margolis, Resnick, & Levy, 2003; Bimber & Davis, 2003; Stanyer, 2010). A meta-analysis found evidence for the existence of both normalisation and equalisation in election campaigns (Strandberg, 2008). The occurrence of either process can differ per country and is dependent on several contextual factors, which will be discussed later on.

Organisational factors

The elements in this factor are about how campaign leaders perceive campaigning. Do they rely on proven best practices from previous campaigns or is there a culture of innovation? John Padgett and Walter Powell (2012) describe the concept of network folding. Applied to the political realm, this entails the extent to which campaigns employ skilled personnel from non-political sectors and to integrate that expertise into their existing institutions. An example is the hiring of Google engineer Stephanie Hannon as chief technology officer by the Clinton campaign (Easton, 2015). The 'cognitive diversity' following from network folding can lead to creative ideas (De Vaan, Stark, & Vedres, 2015). Furthermore, the organisational structure can be expected to resemble the way the campaign perceives PBT. A campaign with an autonomous data department is probably more prone to rolling out a PBT-strategy than a campaign that sees 'data' as only one of the many tasks of a communication staffer. Also, a change in leadership can be a facilitator for innovation (Gibson & Römmele, 2001).

Infrastructural factors

Elements are the technological tools available to campaigns, which enable them to roll out a PBT-operation. For instance, such tools might assist volunteers in the field by enabling them to collect data. They can be developed in-house or outsourced; in fact, there are specialised third party consultancies, that offer off-the-shelf tools, which in turn allow campaigns to employ innovative technology even though the campaigners do not have any technical expertise.

Structural electoral factors1

The actions of rival campaigns fall under the umbrella of structural electoral factors. A successful PBT-campaign of a rival can facilitate innovation in other campaigns, especially if those other campaigns themselves look back at an unsuccessful election. This connects with the 'critical event' (Kreiss, 2016), such as losing an election that should have been won, or with the experience of an 'external shock', which can be an incentive for professionalisation (Gibson & Römmele, 2001).

A second element influencing campaigns' likelihood to use PBT-techniques, is issue ownership (Petrocik, 1996), and the subsequent statements of party candidates propagating standpoints of the party. A political campaign 'caught' using privacy-infringing PBT-techniques, while its candidates present themselves as privacy champions, is likely to come across as hypocritical. Being perceived as such should be avoided, considering the negative electoral consequences following political-ideological hypocrisy perpetrated by politicians (Bhatti, Hansen, & Olsen, 2013).

Ethical and legal concerns

Elements within this factor consist of ethical and legal restrictions on how campaigns operate. For example, a political party could believe that PBT is ethically wrong as it infringes on citizens' right to privacy, and citizens' autonomy to form their own opinions. As a result, the party 'self-regulates' and refrains from using campaigning techniques violating its ethical beliefs.

Another element is the legal uncertainty that occurs when a campaign does not know how to behave in accordance with data protection and election laws, because of a lack of internal expertise. Such confusion can result in differences in the actions taken by comparable actors (e.g. Raskolnikov, 2017). Legal uncertainty can lead to 'overcompliance', which can be seen as a barrier towards the adoption of PBT-techniques, or to 'undercompliance', which facilitates the adoption of PBT-techniques (Calfee & Craswell, 1984). For instance, Anstead (2017) notes how parties felt disadvantaged by targeting possibilities facilitated by the perceived undercompliance with UK campaign finance law during the 2015 general elections.

The system level

Aside from campaign level factors, we look at contextual factors as well. These factors may limit the extent to which (US-American) campaigning techniques can be adopted in other countries (Karlsen, 2010). Therefore, we add three new contextual factors to our model. We expect that the electoral system, the regulatory system, and the culture of a democracy influence the extent to which the campaign team level factors are applicable. Below, we explore how the adoption of PBT-techniques can be influenced by properties of different systems. We will later apply our model (see Figure 1) to one specific case.

Electoral system

The three dominant electoral systems are first-past-the-post (FPTP), proportional representation (PR), and two-round (TR) (Birch, 2001, 2003). How these systems function, can influence how campaigns are run. The FPTP-system, first, can lead to an overvaluation of some key districts. Such districts sometimes 'swing' to one party and sometimes to another party, whereas other districts go to the same party in each election. As an effect, campaigns in a FPTP-system are inclined to spend a disproportionate amount of money and labour in these key districts in the hope of swinging the election their way (e.g., Anstead, 2017; Lipsitz, 2004). The PR-system, second, does not favour a select group of voters in a few key districts (Plasser & Plasser, 2002). This is especially true when the PR-system consists of only one district, in which every vote counts equally. As a result, campaigns have to spread their means more equally over the country. The TR-system, third, makes for a relatively unpredictable campaign, since it often is unclear which candidates will make it to the second round. Furthermore, the TR-system makes it important for campaigns to collect the votes of the supporters of the losing candidates of the first round. Therefore, campaigns should not only focus on their own base but other candidates' bases as well (Blais & Indridason, 2003). This has consequences for PBT, since campaigns should not only correctly classify potential voters as their own, but the other voters as well in order to target them in the next round.

A different aspect of electoral systems that influences how a campaign is run, is the degree of fractionalisation in a democracy (Duverger, 1959; Lijphart, 2012; Wang, 2012). FPTP-systems favour relatively few candidates/parties. PR-systems, in contrast, enable a large number of parties to run in an election. The first round of a TR-system can consist of many different candidates. As a result, campaigns that operate in a PR or a TR-system are less likely to launch attack campaigns against competitors. This is because PR-systems generally require a coalition of parties working together after the elections (Plasser & Plasser, 2002). And in a TR-system, campaigns should not mistreat competing candidates too much because winning campaigns have to court the bases of losing candidates in order to win the second round. Furthermore, in a highly fractionalised democracy, parties represent different (minority) groups within the electorate. This results in a high risk of 'mistargeting', in which campaigns approach a member of group A with appeals for a member of group B. Mistargeting can lead to voters penalising the campaign for their mistake (Hersh, 2013). These contextual circumstances may call for different PBT-strategies.

Regulatory framework

We distinguish strictly regulated campaign environments, moderately regulated campaign environments and minimally regulated campaign environments (Plasser & Plasser, 2002). Strictly regulated environments are characterised by "severe restrictions on the contact and communication between candidates and their constituencies" (p. 137). Moderately regulated environments typically focus on regulating access to TV-advertising and campaign funding. Minimally regulated environments impose few regulatory restrictions on political campaigns. It may be infeasible to implement innovative PBT-techniques in strictly regulated environments. Legal uncertainty can play a role on a system level too (e.g. because of a gap in the law).

Culture

Differences in the adoption and use of innovative PBT-practices can also be influenced by the culture or tradition in a democracy. For example, turnout culture is important because campaigns operating in countries where turnout is high will focus more on convincing voters on getting out the vote than campaigns operating in a low-turnout culture. In a low-turnout culture, campaigns sometimes target specific groups of voters (e.g. the elderly, who are more likely to turn out) more than other groups (e.g. the poor), who are unlikely to turn out (Herrnson, 2001). The turnout culture can influence the data campaigns collect on someone (and how campaigns tailor their messages), because a campaign message meant to convince someone typically leans on more data than a message meant to mobilise a voter does. Furthermore, cultural norms can dictate the strategy of political campaigns. In Japan, for example, posting dark post attack ads, such as the 'super predator' ad Trump launched against Clinton (Green & Issenberg, 2016), is improbable because of the cultural convention of averting direct conflict (Plasser & Plasser, 2002).

System level context is likely to affect campaign level factors. A campaign operating in a multiparty PR-system needs to pour more resources into identifying potential supporters than a campaign in a FPTP-system. After all, identifying potential Republicans or Democrats is easier than identifying potential voters in a ten-party race. Moreover, other than in US campaigns, most European campaigns are unable to access voter registration files provided by an electoral register. In many countries citizens can just show up at the voting booth, which means that the whole act of ‘registering’ for voting, as it is the case in the US, does not exist. Since Hersh (2015) has found that voter lists are one of the most valuable pieces of data to US campaigns, this principal unavailability, or non-existence, of such data poses a challenge for the PBT-capabilities of campaigns. But this challenge should by no means imply a PBT-operation in Europe is impossible. We would argue that while the lack of access to voter lists makes it very difficult to achieve the same level of granularity when engaging in PBT as in the US, by using other commercially available or self-collected data, it can be possible to come reasonably close to the desired level (which may be more modest for European campaign leaders than for American ones). The extent to which there is an actual difference between the degree of granularity between US and European campaigns, however, is outside of the scope of this paper, as we focus on perceptions and strategies of campaign leaders.

System level context also affects infrastructure. For instance, should the groundwork be spread equally across the nation, or focused on a number of battleground states? Also, a campaign operating in a heavily regulated context is likely to encounter some legal barriers. For example, because of campaign financing regulations (may influence resource factors), and data protection regulations (infrastructural factors). The absence of regulations, conversely, can facilitate PBT. Cultural context, finally, can influence campaigns' ethical considerations regarding PBT. Campaigns operating in a culture that favours privacy, for example, can be expected to avoid (or use less-invasive) PBT-techniques than campaigns run in a culture in which privacy is less important. In sum, there are several factors, both on campaign and system level, which can form a barrier or facilitate the extent to which campaigns are able to use PBT-techniques and how they use them.

Extending existing research to a European context, we have developed and will apply an improved model (applicable in different electoral contexts) to analyse barriers and facilitators to innovative PBT-practices by political campaigns. As the context of the research case differs from the US, we expect to contribute to the framework and to shed light on how contextual factors influence innovation of political campaigns. Furthermore, in answering our research question, we provide insight into the way political campaigns in a multiparty democracy organise, communicate and innovate. Given these considerations, our key question is: What barriers and facilitators for the adoption and use of PBT-techniques do Dutch political parties perceive?

Method

This study focuses on campaigns in the Netherlands because of the national elections taking place in the research period (15 March 2017), the advanced technological infrastructure (Coy, 2015), and the interesting contextual factors. The Dutch electoral system is one of open list proportional representation (PR), in which all members of parliament come from one nationwide district (Lijphart, 2012). This means that in the Netherlands, every vote counts equally. Moreover, the system of PR (and the very low de-facto threshold) enables a relatively large number of political parties to run in an election. 28 parties participated in the 2017 national election ("Partijen nemen deel", 2017). Of these parties, 13 actually gained a seat in parliament ("Officiële uitslag", 2017).

The Dutch national elections have a relatively high turnout: around 80% in the previous two elections ("Officiële uitslag", 2017). But where the US presidential campaigns can spend hundreds of millions of dollars (Narayanswamy, Cameron, & Gold, 2017), the Dutch campaign with the biggest funds (VVD) has no more than 5 million dollars to spend. And even if the budgets were sufficiently large, the question is whether voter data would be usable for a political campaign. The Dutch data protection law categorises political preference as sensitive personal data. This means that campaigns are only allowed to process such information if the potential voter explicitly gives permission to do so.

Finally, as party membership in the Netherlands steadily decreases, political campaigns can rely less on their members to do labour-intensive tasks (such as canvassing). In 2016, the number of party members of all political parties combined, was at its lowest point since the second world war. Although this number has picked up slightly since, party membership is still quite low ("Membership Dutch parties still low", 2017).

After approval from the ethical committee of the University of Amsterdam, we carried out eight in-depth interviews with campaign leaders. We interviewed 11 campaign leaders, belonging to eight political parties in total (three interviews were double-interviews). In addition to this, we held two background interviews (with one local campaigner for the municipality of Amsterdam and one political consultant offering PBT-services). The eight elite-interviews on average lasted 53 minutes. Two were conducted by phone, the others face-to-face. We took a qualitative research approach for several reasons: the small group of people concerned with the coordination of political campaigns in the Netherlands, the lack of knowledge on this topic in the Netherlands, and because interviewing is a suitable method for understanding the mechanisms behind and perceptions of a phenomenon (Boeije, 2005). As we want to understand how campaigns see PBT, what they are actually doing, and how they perceive possible barriers and facilitators to the adoption and use of PBT-techniques, the interview is a suitable data-collection method. Using an interview guide (see appendix A), we held semi-structured interviews, allowing follow-up questions.

Interviewees

We selected the interviewees via purposive sampling. Campaign leaders qualified for an interview when they had a coordinating role in the campaign and were campaigning for a party that gained at least one seat in the 2012 national parliamentary elections. Eleven campaigns satisfied this second criterion (see Table 1). We contacted interviewees via email, explaining the objective of the study. The interviewees signed an informed consent document before the interview started. We also promised the campaign leaders anonymity, and confidentiality until after election day (15 March 2017). By doing so, we tried to provide the interviewees with a safe environment in which they felt like speaking freely, and without concern of somehow 'leaking' strategic information. Because no information would become public before election day, the risk that interviewees might provide biased information due to a strategic agenda was minimised. Another large advantage of interviewing the campaign leaders before the elections took place, is the prevention of hindsight bias by the interviewees. Unfortunately, we were unable to convince three parties to comply (VVD [right on the political spectrum], PVV [right-wing nationalist party], PvdD [Party for the animals; left-wing]). These three parties were unwilling to cooperate, either because they still found the risk of leaking their strategy too large, or they did not offer an explanation.

Analysis

Using a grounded theory approach, this study has passed four phases: the exploration phase, the specification phase, the reduction phase, and the integration phase (Wester, 1995). In the exploration phase, two background interviews took place (with a campaigner for the municipality of Amsterdam and with a political consultant offering PBT-services). These were coded using AtlasTI, 'tentatively labelling' relevant information (Glaser, 1978). Thereafter, the first interviews with campaign leaders took place. These were transcribed and open-coded. Furthermore, fellow researchers also coded these interviews and discussed the content (peer debriefing). In the next phase, new interviews took place and the data was subject to axial coding. The first dimensions were identified (e.g., what forms a barrier and what facilitates the use of PBT-techniques?). The reduction phase saw the emergence of the core category (innovation). In the integration phase, we completed the conceptual framework, finalised our analysis, and had the campaign leaders approve the quotes used (member checking). This means the campaign leaders agreed with the way they were quoted, and with the publishing of the names of the political parties. Member checking increased the willingness of campaign leaders to cooperate with the study.

Results

We first describe the field: to what extent do campaigns use PBT-techniques? Then we explain differences between parties by focusing on the five campaign level factors concerning the use of PBT (resource, infrastructural, organisational, structural electoral cycle factors, and ethical and legal concerns). Finally, we zoom out to the system level and discuss the influence of contextual factors (electoral system, regulatory framework and culture) on the adoption of PBT-techniques.

PBT in Dutch campaigns

As campaigns in the Netherlands can have recourse to relatively detailed public census data and detailed election results, all campaigns, to some extent, adopt a PBT-approach. Furthermore, Facebook is an important tool for all parties, but the parties differ in how they use Facebook's capabilities. Some parties occasionally post content targeted to broad age groups, while other campaigns frequently post content tailored to more specific groups. Two campaigns stand out, as they have developed their own PBT-tools, which they can use to continuously refine their knowledge of the electorate. We will now use our model to explain the differences between campaigns.

Resource factors

All campaigns cite financial costs as a barrier. Table 2 shows that budgets are modest, and differ between parties.

Source: parties' annual financial reports, on file with authors.

These small budgets form a barrier for the cooperation with expert political consultants (such as Blue State Digital) to enhance their PBT-operations. Parties refer to the financial costs as the main reason not to hire consultants. Campaign leader 1 of the Social Democratic Party (PvdA) explains why he does not work with Blue State Digital (BSD):

Their system is very expensive, that's a factor. And you need the people to carry out the work for you. In an ideal world, such a cooperation would be really cool though."

Liberal democrat party D66 agrees: "because it costs a lot of money and we don't have that kind of money. And if we spend it on a consultant, we can't spend it on the campaign itself."

The same barrier appears when campaigns speak about other technological means, such as canvassing apps, allowing campaigns to directly process information from canvassers. Christian Democrats CDA, for instance, would like such an app. Campaign leader 8: "yes, but that would demand a financial investment that we can't afford."Green Party GroenLinks has a contrasting perspective: "I believe it usually costs around €100,000 to build an app such as our own. (..) We, however, paid our programmers two crates of beer and 40 pizzas." Several facilitators help GroenLinks and also socialist party SP to overcome this barrier of financial costs. First: the personal network of the campaign leader. This facilitator is especially prominent for GroenLinks, where campaign leader 7 employs his own network to optimise the BSD-systems, but also to help him with setting up other parts of the campaign:

We had to adjust it [the BSD system] somewhat for the Netherlands. The people with whom I did so, Swedish folks ... they are simply a little network of people of around my own age, and some people who are a bit older and have already set up a similar campaign in their own countries. A guy who set up the grassroots organization for Trudeau, for example, he's a couple of years older than I am, but I Skype with him to talk about how I should handle certain things.

Campaign leader 7's personal network plays (or at least played)an important role in cheaply setting up technological tools and creating content:

Through the network, I'm aware of the crowdfunding streams for a normal campaign. (..) I'm meeting a friend tomorrow, who has experience with mail flows. (..) I have a network of volunteering writers, poets, freelance journalists who write for us for free. (..) So partly, I just have a good personal network.

However, as PvdA notes, having lots of data is of no use if you don't have the capacity to use it. GroenLinks tries to overcome this barrier by organising their campaigns, to some extent, in a citizen-initiated manner (Gibson, 2015). A citizen-initiated campaign (CIC), devolves "power over core tasks to the grassroots" (p.183). As campaign leader 7 puts it:

Grassroots is about creating an infrastructure to enable as many sympathizers as possible to volunteer as canvassers on a large scale. So voter contact on a large scale, but also – and that's Bernie's [Sanders] lesson – to have places in which a few people make stuff by themselves without us having any control over it. (..) Embracing people's creativity without managing it.

SP has less need of a citizen initiated campaign, because of their relatively large number of active party members. "The big difference [with GroenLinks] is the fact that we already have the volunteers. Many other parties lack the numbers. We have thousands of party members who gladly canvass for us two weekdays and on Saturday as well." [Campaign leader 11, SP].

Infrastructural factors

Having a good infrastructure allows campaigns to actually collect data and send tailored messages. What kind of PBT-infrastructure can parties rely on and how does it facilitate their use of PBT?

All campaigns use the PBT-infrastructure Facebook offers, although some more than others. Nearly all campaigns use its lookalike audiences function to find new potential voters. Campaign leader 8: "we search for profiles of people who look like the ones who’ve already liked our Facebook page, and then serve them with advertisements." Campaigns also look at people who like pages that are close to the values of the political parties. Christian party ChristenUnie, for instance, tries to target voters who like the page of evangelical broadcaster EO. So does the Calvinist Political Party SGP, which tries to find out people’s interests on Facebook: "For example… farming, or Israel, off the top of my head; you try to approach people along the lines of their interest, or the region in which they reside." [Campaign leader 9]

Some campaigns also employ ‘dark posts’, a Facebook function that enables campaigns to opaquely target specific audiences, while its messages are not visible to untargeted Facebook users. Campaign leader 1 exemplifies:

We’ve managed to get something done related to gas extraction in Groningen. It doesn’t make sense to share that on the national Facebook page, because it was only important news locally. So we put out a dark post, only for Groningen residents. Sometimes we can specify it even more.”

Using Facebook for PBT-purposes, campaigns do not actually gather or own data themselves. There are a few campaigns that do gather their own data, by using canvassing apps. Campaign leader 7:

We use the election results per voting location and use that information to establish the GroenLinks mindedness of a neighbourhood. Then we can prioritise which addresses to visit and which to ignore. When we visit addresses, our volunteers use the app to answer the following questions: 1. Is anyone home? 2. Does she want to talk? 3. Is she going to vote? 4. Is she planning to vote for GroenLinks? 5. What is the most important theme to her? 6. How GroenLinks minded was she? If she considers to vote for GroenLinks, two questions follow: 1. Do you want to stay informed of our campaign by e-mail? 2. Can I have your phone number, so we can ask you to do canvassing talks?”

The GroenLinks app facilitates large scale collection of information about people's political preferences, thereby informing strategic decisions. Also, the personal data can facilitate accurate PBT on an individual level. The secondary objective of the app is to provide an infrastructure for volunteers to campaign on their own terms, whenever they feel like doing so:

Our app, built by hackers, enables others to campaign for us. (..) Someone in [small town] Lutjebroek can install our app and go ahead and work for our campaign. No campaign leader needed. [Campaign leader 7]

Some campaigns monitor the visitors of their own websites. Campaign leader 1: “What are people searching for on our website, how do they get to our website, how much time do they spend, (..) which button should you colour red? How does that work?” At the time of the interview, CDA was not yet tracking their website visitors, but: “we’ve just migrated to a new website, on which we want to start collecting more data on our visitors. I’m curious what kinds of people are visiting the website. And what kinds of people don’t, and therefore have to be reached through different channels.”

SP has built a system which combines previous election results, census data and their own membership Constituent Relationship Management (CRM) data. Plotted on a Google Map, they can identify interesting areas for them to canvass. This system facilitates efficient use of means:

We would do nothing more happily than knocking on every single door in every city, but unfortunately, we do not yet have that kind of manpower. So we do an analysis: What kinds of neighbourhoods are especially interesting for us? We have built our own system to help us make that decision [Campaign leader 11].

Organisational factors

Circumstances within the campaign’s organisation itself can form a barrier for the uptake of PBT-techniques. Less innovative parties, for instance, do not have a dedicated data, tech, or digital department. As a campaign leader notes: "The department responsible for that [tech/data/digital] is our Communication department. So that's four or five people. And sometimes someone of the department picks it up, but there's not one specific person who's responsible."This contrasts with GroenLinks, which has a Digital and Grassroots department and with SP's Digital department.

The "state of mind"within a campaign can also be seen as a barrier:"In the sense that internally, people are still very much inclined to think offline. The culture within the campaign is quite offline." [anonymous campaign leader]2

New leadership and younger staffers can play a facilitating role in political organisations. Campaign leader 11 argues that, because he is young, their new party chair brings a more tech-savvy vision than his predecessor. According to the campaign leader, younger staffers are more likely to implement tech and data in their work procedures.

A final organisational barrier is the primary goal a political party pursues. Campaign leader 10:

Maybe the strange thing about SGP is that we do not care that much about seat maximization. For us, it's about the impact of our principles. And sure, we would rather have four seats than three, but if we have to settle for three seats: that's fine too. And that's, in my opinion, a reason why we have a feeling like: do we really need data?

Structural electoral cycle factors

These circumstances are largely beyond the control of the campaigns, but they can influence the uptake of PBT-techniques. Campaign leaders see the PBT-actions of other political campaigns as a motivational factor. As campaign leader 11 notes about the development of their app: "I've looked a little bit at how GroenLinks have their app and canvassing system." Or as campaign leader 2 concludes: "If every party does it, you don't win very much by it. But if you're the only party that does nothing..."

Ethical and legal concerns

Especially D66 and the seniors’ party 50PLUS take a principled stance against the collection of data and the use of PBT. Where D66 presents itself as a privacy champion and therefore will never gather and use information about (groups of) voters, 50PLUS campaign leader 6 warns about the risk of irresponsible use of the data gathered by the "almost stalking of people", which he calls "morally irresponsible".

Furthermore, a lack of internal legal expertise appears to contribute to a feeling of legal uncertainty, which affects the likelihood of adopting PBT-techniques: "Legislation has grown so very comprehensive and complex. It's almost impossible to cope for us as a small organisation." [Campaign leader 5]

While ethical and legal concerns can form a barrier, a left- or right-wing orientation does not seem to be instrumental therein. After all, we have seen left-wing parties GroenLinks and SP develop relatively advanced PBT-tools. And we have seen right-of-centre party CDA express clear interest in advancing their own PBT capabilities. At the same time, left-of-centre 50PLUS and right-of-centre D66 both oppose the use of PBT.

System level

Electoral system

Although the Dutch one district PR-system should make for a rather equal distribution of campaign efforts, campaigns still divide the country into smaller areas of interest called 'key areas'. These areas differ per party, but do receive a relatively large part of campaign attention. Campaign leader 1 describes these as areas: "where we know the turnout is low, but the number of PvdA-voters is high". All campaigns use data provided by the Electoral Council, showing the election results per party, per voting location to establish key areas. Campaign leader 3 explains:

Using that [the election results], you see: Okay, we do well in this neighbourhood or this street. And then you combine that information with the CBS3 data, to find out what kind of neighbourhood it is, what kind of people live there, what are their backgrounds, how much do they earn, what does the family composition look like, et cetera.

Facilitated by these public data, campaigns enrich their knowledge of specific areas. A next step would be to use those data to make personalised appeals to (subgroups of) people living in those specific key areas.

Regulatory framework

Although the Netherlands would qualify as a minimally regulated environment (Plasser & Plasser, 2002; Esser & Strömbäck, 2012), campaigns all experience regulatory pressure and legal uncertainty on a system level. They cite an abundance of regulations, forming a barrier to their ability to innovate.

The technological developments have been taking place so very quickly. And, in that timeframe, to adjust all your procedures and everything. And also to meet the privacy regulations, I think many parties face a huge challenge in that respect." [Campaign leader 4]

Campaigns sometimes face a dilemma, having to decide between innovative techniques and privacy regulations. Campaign leader 11:

Regulations sometimes are unclear, which leads us to decide to go for the safe option because you do not know where the red line is. And you never want to abuse someone's personal data. So yes, regulations sometimes cause us to hit the brake and that's a good thing.

Culture

There is a recurring worry about the perceived low level of political knowledge of the average Dutch voter. PBT-techniques can facilitate campaigns' efforts to convince or educate such low-information voters, for example by "having a conversation with someone, especially if you share some characteristics," [Campaign leader 7] or by interesting "people for things that are relevant to them and to make them aware of the political dimension of those things." [Campaign leader 8] Campaign leader 3, in contrast, concludes that the electorate's low level of political knowledge (together with the perceived volatility of the electorate, the decrease in political trust, and their focus on persons instead of parties) forms an insurmountable barrier, making PBT-techniques irrelevant.

Discussion and conclusion

In the 2017 elections, used here as a case study, all campaigns use PBT through Facebook, but some parties are more advanced than others, and have even developed their own PBT-tools. We have established what the main barriers and facilitators for PBT are, using five factors on a campaign level and three factors on a system level. Not only does this study shed light on the conditions under which these barriers and facilitators manifest themselves, it also gives insight into their different workings across parties. Our study provides information about the data collected by parties and the PBT-techniques used to attract voters. We demonstrate how personal networks and cognitive diversity within a campaign can level barriers. We show how PBT is not only perceived as useful for campaigns in a FPTP-system, but in a PR-system as well. And we show how regulatory pressure is perceived as an obstacle and as a welcome 'normative red line'.

A triangulated research approach can improve our understanding of the campaign leaders' constructs. Observation of their (use of) PBT-tools and how these tools help campaigns make strategic decisions, can give more insight into the workings of these techniques. Another approach would be to interview canvassers and identify 'field-level' barriers and facilitators. Furthermore, ideally, we would have spoken to all parties holding a seat in parliament. Unfortunately, three parties did not cooperate. Two of those became the largest (VVD) and second largest (PVV; in a very close field) party. Since we did have access to eight of 11 parties, we are confident about our findings and we do not expect to identify additional factors influencing the adoption of PBT from interviews with the remaining parties.

Compared to related recent studies by Anstead (2017), Hersh (2015), Kreiss (2016), and Nielsen (2012), this study makes a number of contributions. In general, we focus our exploratory research on a PR-system instead of a FPTP-system, and we develop a model that takes system level contextual factors into account. Specifically, unlike Anstead (2017) we have found evidence for equalisation (which occurs when smaller parties take advantage of the internet's low costs and direct communication possibilities, and, in doing so, use the internet as a tool of empowerment [e.g. Margolis, Resnick, & Levy, 2003]). This evidence is especially clear in the case of GroenLinks, which was, at the time of the campaign, one of the smallest parties in parliament (now the fifth party). Furthermore, we provide an insightful point of view into Anstead's question of whether "parties develop data-driven capabilities more rapidly in electoral systems with a tendency towards disproportionate outcomes" (2017; p. 23). In comparison with Hersh (2015), we focus less on how differences in data-availability lead to different strategic decisions, and more on how differences in the perception of campaign level and system level factors lead to a variation in the occurrence in PBT-innovation. With regard to Kreiss (2016), we have extended his model and applied it to a multiparty democracy. In comparison with Nielsen (2012), we focus solely on the perception of campaign leaders and not on canvassers. Furthermore, we focus on PBT on online as well as offline platforms.

Our attention for the system level factors has enabled us to identify perceived influence of the PR-system on the adoption of PBT. Contrary to theoretical expectations (Plasser & Plasser, 2002), campaigns in a one-district PR-system do identify key-areas that are more heavily campaigned than other districts. These key-areas differ from 'battleground states' in FPTP-systems in the sense that the key-area does not sometimes swing one way and sometimes the other, but rather that potential voters in key-areas are supportive of a certain party, but not very likely to show up at the polls. Campaigns use PBT-techniques to convince these potential voters of the personal relevance of politics and to motivate them to cast their vote. Areas with firm turnout numbers and clear support for a certain party, in contrast, are perceived as less decisive and less of a priority. This leads to a hierarchy of areas, which differs per party. Also, as a PR-system typically leads to a relatively large number of parties partaking in an election, PBT can be seen as an asset for a campaign to organise in a more efficient manner. Moreover, according to the campaign leaders, PBT-techniques offered by Facebook do allow smaller parties a degree of visibility that they are unable to achieve through traditional media.

On a campaign level, in the coming years, we expect more citizen-initiated campaigning (Gibson, 2015) by campaigns low in labour-resources. This requires a solid infrastructure, which opens the door for third party intermediaries offering off-the-shelf infrastructure. In this regard, it would be interesting to track the development of PvdA, which has suffered its biggest loss in history. This critical event could lead to the prototyping (Kreiss, 2016) of GroenLinks' innovative campaign by PvdA. As the party's chairman has resigned, the door is open to a more cognitive diverse party structure (Du Pre, 2017; De Vaan et al., 2015). Of course, these developments might apply less to parties that are officially more cognisant of campaign ethics (e.g. D66). This is why ethics and legal aspects are important factors to take into consideration. It would be interesting to see how these campaigns act as PBT-capabilities of rival parties improve. Their self-imposed barrier can limit their future chances, but can also attract voters growing more aware of the value of privacy. In the former case, this could lead to an overhaul of their privacy principles, or perhaps to a legislative push towards the restriction of PBT (similar to Hersh, 2015). In the latter case, campaigns can be expected to develop innovative non privacy-invasive campaigning techniques. Either way, our model would provide tools to study the process.

So 'can political parties campaign in Europe as they do in North America' (Bennett, 2016)? We would say ‘mostly yes’. We agree with Bennett (2015) that there are important differences between the US and Europe, and indeed, they influence how PBT is used. But based on our findings, we are hesitant to conclude that those differences (severely) constrain the export of PBT-practices to European multiparty systems. We have shown that relatively small campaign budgets do not need to bar parties from engaging in PBT-practices (or even from cooperating with BSD, an 'expensive' American political consultancy). The same is true of the electoral system: campaign leaders generally perceive PBT-techniques as useful in a PR-system. What remains is the relatively strict Dutch data protection law, labelling political preference as 'sensitive personal data', which can only be processed with explicit consent from the potential voter. 'Explicit consent', however, sounds harsher on paper than it is in practice and is easily achieved (e.g., Beales & Muris, 2008; Calo, 2012; Joergensen, 2014). Of course, because of data regulations and/or their non-existence, European campaigns are unable to consult voting lists showing whether an individual showed up at the polls in the last elections. In most European countries, the electoral register is inaccessible to political parties. One might argue that, from a campaign's perspective, US voter data are superior to European voter data. We would argue that European data are different, but they do not bar European campaigns in the use of PBT-techniques. Dutch campaigns, for instance, can (and do) rely on election results on voting booth level (which comprises a couple of streets). They can (and do) combine these results with detailed, accurate, and a multitude of data about the neighbourhoods surrounding those voting booths. And then there is Facebook, facilitating easy targeting of its users with personalised messages. As potential challenges for democracy come with PBT, such as ignoring 'less valuable' citizens (e.g. reliable non-voters), more research into the workings and effects of PBT is needed.

Disclaimer: This guest essay in the Special issue on political micro-targeting has not been peer reviewed. It is an abbreviated version of a speech delivered by the Member of the European Partiament (MEP) Sophie in ‘t Veld in Amsterdam in May 2017 to Data & Democracy, a conference on political micro-targeting.

Democracy

Democracy is valuable and vulnerable, which is reason enough to remain alert for new developments that can undermine her. In recent months, we have seen enough examples of the growing impact of personal data in campaigns and elections. It is important and urgent for us to publicly debate this development. It is easy to see why we should take action against extremist propaganda of hatemongers aiming to recruit young people for violent acts. But we euphemistically speak of 'fake news' when lies, 'half-truths’, conspiracy theories, and sedition creepily poison public opinion.

The literal meaning of democracy is 'the power of the people'. 'Power' presupposes freedom. Freedom to choose and to decide. Freedom from coercion and pressure. Freedom from manipulation. 'Power' also presupposes knowledge. Knowledge of all facts, aspects, and options. And knowing how to balance them against each other. When freedom and knowledge are restricted, there can be no power.

In a democracy, every individual choice influences society as a whole. Therefore, the common interest is served with everyone's ability to make their choices in complete freedom, and with complete knowledge.

The interests of parties and political candidates who compete for citizen’s votes may differ from that higher interest. They want citizens to see their political advertising, and only theirs, not that of their competitors. Not only do parties and candidates compete for the voter's favour. They contend for his exclusive time and attention as well.

Political targeting

No laws dictate what kind of information a voter should rely on to be able to make the right consideration. For lamb chops, toothpaste, mortgages or cars, for example, it’s mandatory for producers to mention the origin and properties. This enables consumers to make a responsible decision. Providing false information is illegal. All ingredients, properties, and risks have to be mentioned on the label.

Political communication, however, is protected by freedom of speech. Political parties are allowed to use all kinds of sales tricks.

And, of course, campaigns do their utmost and continuously test the limits of the socially acceptable.

Nothing new, so far. There is no holding back in getting the voters to cast their vote on your party or your candidate. From temptation with attractive promises, to outright bribery. From applying pressure to straightforward intimidation.

Important therein is how and where you can reach the voter. In the old days it was easy: Catholics were told on Sundays in church that they had no other choice in the voting booth than the catholic choice. And no righteous Catholic dared to think about voting differently. At home, the father told the mother how to vote. The children received their political preference from home and from school. Catholics learned about current affairs via a catholic newspaper, and through the catholic radio broadcaster. In the Dutch society, which consisted of a few of such pillars, one was only offered the opinions of one's own pillar1. A kind of filter bubble avant la lettre.

Political micro-targeting

Nowadays, political parties have a different approach. With new technologies, the sky is the limit.

Increasingly advanced techniques allow the mapping of voter preferences, activities, and connections. Using endless amounts of personal data, any individual on earth can be reconstructed in detail. Not only can their personal beliefs be distilled from large troves of data, no, it even is possible to predict a person's beliefs, even before they have formed them themselves. And, subsequently, it is possible to subtly steer those beliefs, while leaving the person thinking they made their decision all by themselves.

As often is the case, the Americans lead in the use of new techniques. While we Europeans, touchingly old-fashioned knock on doors and hand out flyers at Saturday's market, the American employ the latest technology to identify, approach, and influence voters.

Of course, trying to find out where voters can be reached and how they can be influenced is no novelty. Political parties map which neighbourhoods predominantly vote for them, which neighbourhoods have potential, and in which neighbourhoods campaigning would be a wasted effort. Parties work with detailed profiles and target audiences, for which they can tailor their messages.

But the usage of personal data on a large scale has a lot more to offer. Obviously, this is a big opportunity for political parties, and for anyone else, who runs campaigns or aims to influence the elections.

However, the influencing techniques become increasingly opaque. As a result of the alleged filter bubble, voters are being reaffirmed in their own beliefs, and they hardly receive information anymore about the beliefs and arguments of other groups. This new kind of segmentation may stifle critical thinking. There may not be enough incentive to test one's own ideas, to find new arguments, or to critically reflect on the truthfulness of information.

I am a social and economic liberal D66 politician, and I get suggestions for news articles from websites like The Guardian or Le Monde. My colleague from the right wing nationalist PVV, may well receive URLs from Breitbart.

Pluralism is essential for a healthy, robust democracy. In a polarised society, people live in tightly knit groups, which hardly communicate with each other. In a pluralist society people engage in the free exchange, confrontation, and fusion of ideas.

The concept pluralism is under pressure. Populist parties declare themselves representative of The People. In their vision, The People, is uniform and homogenous. There is a dominant cultural norm, dictated from the top-down, to which everyone must conform. Whomever refuses, gets chewed out. Often, it is about one-dimensional symbolism such as Easter eggs and Christmas trees. There is no place for pluralism in the world of the populists. But when there is no pluralism, there is no democracy. Without pluralism, democracy is nothing more than a simple tribal dispute, instead of the expression of the will of all citizens together.

Voter data

European privacy legislation limits the use of personal data. In the world of ‘big data’, one of the explicit goals of regulation is to prevent restriction of the consumer's choice. Oddly enough, lawmakers do not explicitly aspire to guarantee voters as broad a choice as possible. But in politics, individual choices have consequences for society as a whole.

In 2018, the General Data Protection Regulation (GDPR) comes into effect. We have worked five years on the GDPR. At this moment, we work on the modernisation of the e-Privacy Directive, which is mainly about the protection of communication. As was the case with the GDPR, companies from certain sectors scream bloody murder. European privacy protection would mean certain death for the European industry. According to some corporate Cassandras, entire European industries will move to other continents. That very same death of corporate Europe is also predicted for any measure concerning, say, environmental norms, procurement rules, or employee rights. All those measures are in place, but, as far as I know, the nightmare scenario has never occurred...

There are some corporate sectors, such as publishing and marketing, which have a huge impact on the information supply to citizens. They are the ones who now cry wolf. It is understandable that they are unhappy with stricter rules concerning their activities, but as the potential impact of the use of personal data and ‘big data’ increases, so does their social responsibility.

At the moment, there is not much public debate about the new techniques. Peculiar. Thirty years ago, 'subliminal advertising', as we called it then, was prohibited because people found it unethical to influence people without their knowledge. We need to have a similar debate. What do we think of opaque influencing? Do we need ethical norms? Should such norms apply only to political campaigns, or should we look at this from a broader perspective? In the ‘big data’ debate, we tend to speak in technical or legal terms, while actually the issue is fundamentally ethical, holding far-reaching consequences for the vitality of our democracy.

Such a public debate demands more clarity on the impact of ‘big data’, profiling, targeting, and similar techniques on the individual, her behaviour, and her choices, which determine in what direction society progresses. Which voters are being reached? How sensitive are they for the subtle influencing and what makes them resilient? How do people who are hardly reached only compare to the others? How do voters and non-voters compare? Is the voter truly predictable? Can we identify or influence the floating voter? Do voters actually float between different parties? Or do they especially float within their own party, their own bubble, their own segment? How important are other factors, such as the social context? If the new influencing techniques are indeed as potent as we think, how can polls get it so wrong? What can we learn from advertisers who return to contextual advertising, because targeting turns out less effective than they thought?

We need to stay cool-headed. New technologies have a huge impact, but human nature will not suddenly change due to ‘big data’ and its use. Our natural instincts and reflexes will definitely not evolve in a few years. That would take many thousands of years, as even in the 21st century, we seem to have more than a few cavemen traits, so losing internalised behaviour is not as easy as 1-2-3. Humans are resilient, but democracy is vulnerable. On a short term, the societal impact is large. This gives us all the reason to reflect on how to deal with the new reality, and how we can keep up our values in this new reality.

The use of personal data, clearly, is not solely reserved for decent political parties. Other persons and organisations, from the Kremlin to Breitbart, can bombard European voters with information and misinformation. But European governments, controlling endless amounts of personal data of their citizens, can also manipulate information, or circulate utter nonsense to advance their own interests. A random example: the Hungarian government influencing their voters with lies and manipulation about the so-called consultation on asylum seekers.

Beyond voter data

This issue is not only about the personal data of voters, but also about the personal data of political competitors, opponents, and critics, which are increasingly being employed. Recently, we have seen efforts of external parties to influence the results of the 2017 French elections. We saw a large-scale hack of the Emmanuel Macron campaign, and the spread of false information, coming obviously from the Kremlin and the American Alt-Right, meant to discredit Macron's candidacy.

Also, the American elections show the shady game of hacking, leaking, and manipulating. The issue of the Hillary Clinton mails will undoubtedly occupy our minds for years. Who knows how the elections would have turned out without this affair?

Other democratic pillars can get corrupted as well by the misuse of data. Critical voices, opposition, and checks and balances are democracy's oxygen. Democracy is in acute jeopardy when data are employed to attack, undermine, discredit, blackmail, or persecute journalists, judges, lawyers, NGOs, whistleblowers, and opposition parties.

In Europe, we tend to shrug our shoulders at these dangers. "Oh well, we'll see, such things occur only in banana republics, not right here". Of course, this trust in our democratic rule of law is wonderful. But if we treat our rule of law this neglectfully, we will lose it eventually.

Within the European Union, we currently see this happening in Poland and Hungary. The governments of both nations ruthlessly attack independent judges, critical media, inconvenient NGOs. They do so with quasi-lawful means. Under the banner of transparency, they force NGOs to register. In doing so, they misuse laws against money laundering, and terror finance. Or the governments bring out compromising information about judges or politicians in strategic moments.

But critical voices struggle in other member states as well. Lawyers are being monitored, even without a legal basis. In the years after 9/11, we have created endless new abilities for intelligence services, police and justice departments to spy on citizens, even without suspicion, without the signature of a judge. The companies to which we unwittingly surrender our personal data, in exchange for service, are forced to hand over all information to the government, or forced to build in backdoors. Governments hack computers in other countries. Usually, it starts out with unlawful practices, but soon enough laws are put in place to legalise those practices. The magic word 'terrorism' silences any critique on such legislation.

But when politicians, journalists, NGOs, whistleblowers, lawyers, and many others cannot perform their tasks freely and without worry, our democracy withers. Not only do they have to operate without someone keeping an eye on them, they have to know nobody is in fact watching them. The mere possibility of being watched, results in a chilling effect.

For this principal reason, I have contested a French mass surveillance law before the French Conseil d'Etat. Since, as a member of the European Parliament, I spend four days a month on French soil (in Strasbourg), I could potentially be the target of the French eavesdropping programme. This is not totally imaginary, as I am not only a politician, but also a vocal critic of certain French anti-terror measures. It is not about me actually worrying about being spied on, but about the fact that I might be spied on. Luckily, I am not easily startled, but I can imagine that many politicians are vulnerable. That is a risk for democracy.

I do not discard the possibility of a ruling of the European Court of Human Rights on my case. In that turn of events, it will lead to jurisprudence valid in the entire EU (and the geographical area covered by the Council of Europe).

But, of course, this should not depend on the actions of one obstinate individual whether politicians, NGOs, journalists, and so on, can do their jobs fearlessly, to fulfil their watchdog role.

It is my personal, deep, conviction that the biggest threat to our democracy is the fact that we have enabled the powerful to access, with almost no limitations, the personal data of those who should control those very same powerful entities.

What can we do?

Some propose new forms of democracy, in which universal suffrage is weakened or even abolished. In his book ‘Against elections: the case for democracy’, David Van Reybrouck had the idea to appoint representatives on the basis of chance, and in his book ‘Against democracy’ Jason Brennan wants to give the elite more votes than the lower classes, presuming that people with more education or development make better choices. Others want to replace representative democracy with direct democracy.

I oppose those ideas. Universal suffrage and the representative democracy are great achievements, which have led to enormous progress in society.

First of all, we have to make sure our children grow up to be critical, independent thinkers. Think differently, deviate, provoke: this must be encouraged instead of condemned. A democracy needs non-conformists.

We must teach our children to contextualise information and to compare sources.

The counterpart of ‘big data’ must be ‘big transparency’. We need to understand not just open administration, but also insights into the techniques of influence.

The regulation and limitation of the use of personal data, as I hope to have argued effectively, is not a game of out-of-touch privacy activists. It is essential for democracy. We need safeguards, not only to be sure people really are free in their choices, but also to protect the necessary checks and balances. As such, I plea for a rigorous application of the GDPR, and in the European Parliament, I will work for a firm e-Privacy Directive.

And yes, perhaps we should examine whether the rules for political campaigning are still up-to-date. In most countries, those rules cover a cap on campaign expenditures, a prohibition of campaigning or polling on the day before election day, or a ban on publishing information that may influence the election results, such as the leaked e-mails in France. But these rules have little impact on the use of personal data to subtly influence elections.

Last year, the European Parliament supported my proposal for a mechanism to guard democracy, the rule of law, and fundamental rights in Europe.2

On this day (editor’s note: 9 May, Europe Day) of European democracy, I plead for equal, high norms in Europe. The last years have shown that national elections are European elections. It is crucial for us to trust that all elections in EU member states are open, free, and honest elections, free of improper influencing.

These last sixty years, the European Union has developed itself into a world leader in democracy and freedom. If we start a public debate, Europe can remain a world leader.

The California Consumer Privacy Act (CCPA), slated to enter into force on 1 January 2020, borrows some cutting edge ideas from the EU and others’ privacy regimes while also experimenting with new approaches to data privacy. Importantly, the CCPA envisages an online advertisement market in which business are prevented from “getting high on information,” 1 breaches are promptly notified, and consumers are autonomous participants with the ability to sell their data at will. Where the CCPA breaks new ground is in protecting consumers from retaliation for opting out of the sale of their data. Thus, if it lives up to its potential, the CCPA could catalyse a permanent restructuring of the online data mining business. Our contribution will shed light on the new CCPA and offer some observations in comparing it with EU’s General Data Protection Regulation (GDPR).

The federal data privacy inerTia

“Try to steal your mind's elation . . . dream of silver screen quotation”- Red Hot Chili Peppers

In spite of recent grumblings in political Washington following the Cambridge Analytica scandal there is for the time being no expectation that the US government will pass comprehensive federal legislation governing consumers’ data privacy. Political willpower to regulate the societal harms of technology at the federal level has long been stalled. Since the 1995 closure of the Office of Technology Assessment, there has been a void in federal oversight over the online collection and sale of data.2 However, over the last decade, the FTC has stepped in to fill the legislative void by investigating data breaches and privacy concerns. 3

However, informational privacy is gaining increasing momentum. Recently, the National Institute of Standards and Technology (NIST), an agency housed within the Department of Commerce, announced plans to develop federal guidelines for the protection of consumers’ online data. In early October, 2018, Democratic Congressman Ro Khanna, who represents Silicon Valley in the House of Representatives unveiled his "Internet Bill of Rights", a blueprint for federal regulation, which would oblige tech firms to alert users of data breaches, provide disclosure, consent and portability inter alia. If such a federal law were passed, it would override, or preempt state regulations, however given deep partisan divisions and the slow pace of regulatory innovation in Washington, the odds of swift federal action currently appear dim.

As Supreme Court Justice Louis Brandeis aptly pointed out in 1932, the states are “the laboratories of democracy”. While US senators have been grandstanding – and maintaining the status quo -- California has been early on experimenting with data privacy rules. As retold in a New York Times magazine feature, the creation of California's CCPA is an example of strategic advocacy for change. In 2018, a group of California residents, shaken by the recent Cambridge Analytica scandal, pressured legislators in Sacramento to pass a law that would give California consumers more control over their personal data. 4 However, rather than put the law into the hands of voters who might have demanded an even more radical restructuring of the online ad industry, the legislature chose instead to negotiate its own version of AB 375 and signed it into law on 28 June 2018. The new law provides a huge leap forward for California’s citizens and is considered a game-changer for the nation, creating justiciable rights vested in both consumers and the California Attorney General.

The ‘California effect’

And if you want these kind of dreams . . . It’s Californication (Red Hot Chili Peppers)

If the state is indeed a laboratory for new ideas, the impact of California’s data privacy experiment may be monumental given the state’s status as the home to Silicon Valley, one of the world’s largest economies, and the second largest state in the US. The positive ripple effect to be gained from California’s more stringent standards, what David Vogel has coined the ‘California effect,’ refers to the upwards pull higher regulatory standards can exert on an industry as a whole. 5 Researchers, such as Bilyana Petkova, contend that federalism or (state jurisdiction) can create “races to the top” in data privacy by enabling states to act sooner than the federal government and to cross-pollinate among state jurisdictions. Also, the testing of innovative policies at the state-level allows federal legislators to later adopt policies that were particularly successful at the state level.

Even prior to the passage of the CCPA, California has already been a frontrunner in driving certain particular data privacy policies. In 1972, Golden State voters amended their constitution to include the right to privacy as an “inalienable right”. In 2003, the California Online Privacy Protection Act (CalOPPA) became the first major consumer data handling law in the United States. CalOPPA focused on the availability of privacy policies, compliance with protective Do Not Track (DNT) settings, and the safe storage of consumers’ data. In 2005, the “Shine the Light” law was enacted to protect California consumers’ private information by requiring transparency and the disclosure of the identity of third parties receiving consumers’ data. 6

In a sign of its responsiveness to the evolving privacy needs of the digital marketplace, California has taken the lead among US states in regulating the mobile app economy. In 2012, the Attorney General of California and six leading app stores struck an agreement to strengthen privacy protections on mobile apps. By ensuring the inclusion of privacy policies in mobile apps catering to a California audience, the mobile privacy deal was credited with creating positive ripple effects beyond California and throughout the app ecosystem. Similarly, the CCPA may also effectively raise the bar for data privacy protection for non-California residents.

California’s electronic jurisdiction

Space may be the final frontier but it's made in a Hollywood basement” (Red Hot Chili Peppers)

The CCPA protects only California residents (humans, not companies) and solely in their role as consumers in a commercial setting. 7 Like the European GDPR, the California law relies on the underlying logic that individuals’ data should be protected according to the whereabouts of the individual supplying the data. Thus, the CCPA applies to businesses around the whole world as long as they reach out to California residents who are present in their state. This means, for example, a California resident, living in Los Angeles whose data is sold by a business in India to a business in Canada would have justiciable rights under the CCPA.

Whilst the CCPA only gives protection to California residents, in practice, the CCPA’s California jurisdiction is likely to raise the bar for users outside of California. The CCPA explicitly states that businesses are free to market their online goods and services to non-California residents without providing opt-out buttons or abiding by the CCPA’s disclosure requirements. However, it will be difficult to disambiguate data attributable to California residents from residents of other states. The administration cost associated with maintaining separate websites for California traffic is likely to cause many businesses to simply raise their data privacy standards for all users.

The scope of CCPA

Personal data: The CCPA gives as broad a coverage to personal information as the GDPR. Any information capable of being linked indirectly or directly to a particular consumer or household is protected. Also similar to the GDPR, the CCPA covers both electronic and offline data physically given to companies.

Businesses covered: The CCPA is intended to govern heavy hitters: it applies to big data brokers, not small and medium size enterprises and nonprofits. CCPA regulated businesses must be for-profit firms either making over $25 million in annual revenue, or businesses holding the personal data of 50,000 people, households, or devices; or businesses that gain at least half of their revenue from the sale of personal data.

Business purpose exception: Most internal collection and use of data can be shielded from opt-out under the so-called business purpose exception. Generally, businesses need not seek consent for one-time transactions such as payments with a credit card and interactions in which the data is not sold or retained or where the data collected does not meet the definition of personal data. Also, the CCPA does not envision an end date for the internal use of data, enabling businesses to hold consumers’ personal information indefinitely. By contrast the GDPR regulates all instances of personal data processing irrespective of whether it is internal to an organisation’s business or involves personal data shared with third parties. However, as explained below, CCPA protected consumers will have the right to disclosure of their data collected for a business purpose.

CCPA’s impact

This is what you’re craving . . . Californication (Red Hot Chili Peppers)

Gone are the days when California consumers unwittingly forked over their personal information to powerful tech platforms only to lose track of their data when it was sold further on. The CCPA empowers consumers to say no to the sale of their own data. By contrast, the EU’s GDPR approaches data privacy from a comprehensive human rights framework, offering a high level of protection throughout the lifecycle of residents’ personal data from the moment of its collection, in relation to each use and until its eventual deletion. Still the CCPA improves transparency, allowing consumers to obtain a portable copy of their data and to discover who is holding which of their data, and for what purpose.

One feature unique to the CCPA is a provision that protects consumers from retaliation if they opt out of the sale of their data. For instance, prior to the CCPA, a person wishing to use online map navigation could be left stranded if they declined the business full access to their personal data. After the CCPA, businesses may no longer downgrade the level and quality of services simply because a consumer has opted out. Another exciting innovation is the law’s financial incentives that encourage businesses to compensate consumers who wish to sell their data.

The major market shift brought by the CCPA resides in the autonomy it restores to consumers. California consumers will now have full ‘ownership’ over their own data, with the right to sell their data should they so choose. Before the advent of the CCPA, consumers eager to access the basic infrastructure of the online world had to sign away their data without even realising the value of the personal data they gave up. That dynamic changes radically under the weight of the CCPA, which specifically mentions the right of businesses to pay for the collection of consumers’ personal information. In sum, while platforms are prohibited from downgrading consumers, the CCPA encourages practices that allow consumers to monetise their data. This provision is a laudable step toward strengthening consumers’ bargaining power vis-à-vis platforms and raising awareness among consumers of the existence of the data market.

The nuts and bolts of CCPA

It’s understood that Hollywood / Sells Californication (Red Hot Chili Peppers)

What follows is a primer on CCPA’s most prominent features, which taken together, aim to restructure power asymmetries in the collection, sharing and control of personal information.

Verification of consumer identity: The CCPA requires that only verifiable requests for the disclosure of a consumer’s data be honoured. In practice verifying a consumer’s identity will not be very complex. The CCPA is generous toward consumers in its definition of verifiable requests. Businesses must treat requests coming from password-protected accounts maintained by a consumer and made while the consumer is logged-in as verifiable.

Data requests: Somewhat mirroring the EU’s right to access, California consumers may in 2020 request a record of the types of data a business holds on them, including information about “business use” and third-party sharing of their data. Businesses are to provide portable information that can be readily transmitted to another business.

Disclosure: Businesses will have to inform consumers of the data they are collecting from their interactions. Disclosure must include the types of data sold, purpose for its use, and the identity of the third party recipients. Notably, while the business purpose exception precludes consumers from controlling businesses internal processing of their data, the disclosure provision does allow consumers to discover which of their data businesses have collected and stored for a business purpose.

Right to ‘Deletion’: Akin to Europe’s “right to be forgotten,” California consumers may now request erasure of their information. Businesses must inform consumers of this right. This right, in combination with the disclosure of data held for internal use, may serve as a strategic counterweight to the business purpose exception. Consumers may request to know which of their personal data a business is storing and then request to have it deleted. However, businesses may use a number of exceptions to escape compliance with deletion requests, including completion of a transaction, cybersecurity, debugging, research, free speech, and some internal analytical use. Sadly, the analytical use exception could make it extremely difficult for a consumer to get personal data stored for internal use deleted.

‘Right to Say No’ to data sale (opt-out): In a dramatic shift, businesses will now have to obtain consent from consumers -- at or before collection – for the sale of their personal information. Under the CCPA, consumers may, at any time, exercise their right to decline consent to the sale of their personal information. Importantly, the CCPA appears to remedy much of the shortcomings of consent boxes by requiring businesses to place a "Do Not Sell My Personal Information" button on their websites. The button may be a good fix for design issues with opt-out box ticking.

Children’s data: Furthermore, the sale of children's data will require express opt in, either by the child, if between ages 13 and 16, or by the parent if younger than that. 

Third party liability: The CCPA creates third party liability when “service providers” and “third parties” processing data violate consumers’ data privacy by failing to obtain consent for commercial sale of data or failing to make disclosures of data breaches. This could become a very important measure to counter the murky practice of third party tracking online.

Sanctions: While the CCPA is strong on consumer autonomy, it seems weak on statutory relief for individual consumers.Failure to address an alleged violation within 30 days could lead to a $750 fine per consumer and per incident or actual damages, whichever is greater. This $750 per person fine seems laughable compared to the GDPR’s whopping 4% of global turnover. Individuals are unlikely to exercise their right to sue under the CCPA if they only stand to gain $750 as a result. Legislators could have better served individual consumers by creating a higher threshold for fines in an action brought by a single person.

On the other hand, the $750 fine could have a strong deterrent effect on aggregate. Class action suits brought by individuals or the Attorney General could make the CCPA put its money where its mouth is. For instance, if one-third of California’s population of 39.5 million people were subject to an undisclosed data breach, the Attorney General would be able to wrestle back up to about $10 billion. Still, to give the CCPA any teeth, the Attorney General of California will have to settle on a definition of “per violation” that covers each discrete usage of a person’s data per instance. It is important to note, however that the $750 statutory fine is not the only available remedy for a violation of the CCPA. Judges may impose any relief the court deems proper, including punitive damages in amount high enough to deter repetition of the violation in the future.

Non-discrimination for exercise of rights: a big win for consumers

Firstborn unicorn . . . dream of Californication (Red Hot Chili Peppers)

A major problem with data privacy regulation is the idea that businesses may pull back their services under the pretext that compliance is too costly. The CCPA resolves this risk by protecting consumers from being denied goods or services or charged differential prices simply because they have opted-out of the sale of their data. In essence, businesses may not punish consumers with higher prices or lower quality simply because they exercise their rights under the CCPA.

Yet there is a risk that this strong non-derogation guarantee may be subsumed by the exception that follows it. The provision that follows states that businesses may in fact offer variations in level or quality of their online goods and services as long as they can prove that the consumer’s data is germane to the added level of service. Hopefully this exception will be interpreted narrowly or removed entirely. For instance, a consumer who chooses to give a business her phone number may be contacted by the company via telephone to ensure she is happy with the service while a consumer who opts out of sharing this data may not receive customer service over the phone. Clearly this exception, if interpreted too liberally, risks undermining the guarantee that customers will not lose quality if they opt-out. It is therefore foreseeable that this exception will either be amended, scrapped or litigated after 1 January 2020.

Private right of action for consumers

Is it war you’re waging? (Red Hot Chili Peppers)

Consumers may bring individual or class action civil lawsuits under the CCPA in cases of negligent data breach. 8 What looks revolutionary from the outset, unfortunately appears to erect cumbersome barriers to consumer redress. For instance, consumers must wait 30 days to see if a business in violation can “cure” the breach. If within these 30 days, a business is able to show it has remedied the problem and if it provides the consumer an assurance that “no further violations shall occur,” then the consumer must desist. Also, consumers must defer to the Attorney General, to see if the Attorney General wishes to prosecute on its own initiative before filing suit.

It is also unclear how the Attorney General will accomplish the Herculean task of auditing the use of data outsourced to “service providers” under the guise of the business purpose exception. 9 While the Attorney General is a formidable prosecutorial authority, its competence is spread over both criminal and civil law. Furthermore, funding for CCPA prosecution is to come directly out of the fines garnered from successful prosecutions of data privacy violations. It is therefore likely that the Attorney General of California will lack the level of resources allocated to European Data Protection Authorities that have been specialising in data privacy discipline for years.

Conclusion

Destruction leads to a very rough road, but it also breeds creation. (Red Hot Chili Peppers)

By shining a light on the collection and use of data, requiring consent for data sales, and giving greater autonomy to consumers, the CCPA is the most remarkable achievement in US data privacy law to date. The CCPA’s opt-out provision makes consumers captains of their own ship rather than unwitting stowaways at the mercy of powerful platforms. However, the law still contains a number of potential loopholes that need shoring up. For instance, the Attorney General must still address the law’s administrability problems, the need for a sharper definition of business purpose, inadequacy of individual fines, the question of how businesses approach the positioning and prominence of "Do Not Sell My Personal Information" buttons, and vagueness in the non-discrimination protection.

Yet despite these shortcomings, the CCPA overcomes the highest hurdle standing in the way of data privacy: it creates demand for rights among consumers. Consumers cannot take control of their own data if they do not know they are entitled to it. The "Do Not Sell My Personal Information" button cleverly nudges consumers toward exercising their rights by reminding them that their data is in fact being sold at the same time that it instills a sense of ownership over one’s personal data where there previously was none. The law’s incentivising financial compensation in exchange for consumers’ sale of their personal data is also helpful. One can only hope that over time the sense of entitlement to personal data rights becomes entrenched such that consumers refuse to bear the burden of big data’s most pernicious negative externalities.

The CCPA’s proponents must be applauded for accomplishing what until very recently looked like an implausible California dream. The new law is in many aspects weaker than what is mandated under the European Union’s GDPR but it is not creating conflicts of laws for transatlantic businesses. In fact it is an interesting new lab to observe whether the excesses of today’s online tracking panopticon can be squeezed for good between the GDPR and the CCPA. Both instruments share the principle dilemma of leaving consumers to micro-manage their data privacy preferences which is another form of non-gratified labour. This necessarily places individuals at a strategic disadvantage because rights need to be exercised to come to life.

This paper is part of 'A Manchurian candidate or just a dark horse? Towards the next generation of political micro-targeting research’, a Special issue of the Internet Policy Review.

Introduction

As political campaigns compete, they try to outsmart each other by all sorts of actions: from dropping witty puns during a televised debate, to strategically knocking on doors and convincing voters. Technological innovation can help political parties improve the effectiveness of their campaigns. By using technology to collect, process, and analyse information about voters, campaigns can improve their knowledge about the electorate. Subsequently, technology can extend campaigns' capabilities of targeting specific groups with tailored messages resulting in more efficient campaigning. We call this phenomenon 'political behavioural targeting' (PBT).

Several scholars have researched political behavioural targeting in the US context (e.g., Kreiss, 2012, 2016; Nielsen, 2012; Hersh, 2015). However, the US differs in several obvious ways from most European countries. One can imagine that differences in electoral systems, privacy laws, and party financing influence campaigns’ ability to collect, process, and use personal voter data. Therefore, the findings from these studies do not necessarily apply to European countries. As there is little research in a European context, it remains unclear to what extent and how campaigns in a multiparty democracy, such as the Netherlands, use PBT-techniques. Also, it is unclear if and why there are differences between parties. In line with Colin Bennett (2016, p. 261), we wonder: "can political parties campaign in Europe as they do in North America?"

Such a question is relevant, as some scholars fear that the use of data and targeting techniques hinders public deliberation (Gorton, 2016), weakens the mandate of elected officials (Barocas, 2012), has negative effects on citizens’ privacy (Howard, 2006; Rubinstein, 2014; Tene, 2011), and enables campaigns to send tailored messages directly to citizens, thereby avoiding scrutiny from journalists (Jamieson, 2013). As a result, campaigns can potentially make opposite promises to different people, without anyone noticing.

This article sheds light on how Dutch political campaigns adopt and use PBT-techniques. Through interviews with campaign leaders, using a grounded theory approach, we answer the following overarching research question: What barriers and facilitators for the adoption and use of PBT-techniques do Dutch political parties perceive?

Theoretical framework

We will first summarise innovations in political campaigns over time, leading up to the advent of political behavioural targeting. Then, we identify the factors influencing the adoption of PBT on a campaign team level. Finally, we explore the factors that can shape the adoption of PBT on the level of national systems.

Innovations in political campaigns

Political campaigns have continuously been adapting to technological developments. Pippa Norris (2000) describes how the advent of television and the shift from partisan newspapers to national television news triggered a process of modernisation in the way political campaigns operated. Notable consequences of this shift were the adoption of a media-centred strategy in order to set the agenda, the rise of political marketing, the collection and use of data (such as opinion polls) to "shape, fine-tune and monitor campaign efforts" (Blumler, Kavanagh, and Nossiter, 1996; p. 53). Another shift came with the internet and the new possibilities for party-voter interaction that came along with the medium, which led campaigns to a new stage of the modernisation process: the postmodern campaign (Norris, 2000).

It would be an oversimplification to point to 'the internet' as a game-changer in political communication, because of the rapidly changing nature of the internet itself. As David Karpf (2012, p. 640) notes: "the internet of 2002 has important differences from the internet of 2005, or 2009, or 2012". Accordingly, much more than the advent of the internet itself, it is the advent of social media such as Facebook (2004), YouTube (2005) and Twitter (2006) which provided political campaigns with new ways of communication with the electorate (e.g., Gibson & McAllister, 2011; Conway, Kenski, & Wang, 2015; Vaccari, 2012). Together with companies such as Google, whose core business is actually not its well-known search engine but rather its advertisement business, social media not only facilitate new ways of communication, but also the tracking and collection of behavioural data of internet users (Zuiderveen Borgesius, 2016). This technique ("behavioural targeting") originates from the advertisement business. Ad agencies monitor people's online behaviour and combine this information with consumer data provided by data brokers, to target them individually with tailored ads (Turow, 2011, p.75). When applying this concept to the political realm, we can dub this phenomenon as political behavioural targeting (PBT).

Of course, PBT is not about selling products but about winning votes. And political campaigns have different means to do so than advertisement agencies have (e.g. canvassing efforts); which means that PBT happens offline as well as online. We distinguish traditional canvassing from PBT-canvassing if campaigns are able to process information about individual conversations (such as the voter's likelihood to vote for a party or her most important voting consideration), and subsequently use that information to gain strategic insights about the electorate and/or to target the voter at a later stage with a tailored message, while skipping the 'wrong' doors in a neighbourhood (Kreiss, 2016; Nielsen, 2012).

Arguably, the use of PBT can be seen as the latest step within the modernisation of political campaigns. However, as we have seen in earlier phases, not all parties in all countries adopt new techniques at the same pace and rate. Below, we identify the factors influencing the adoption of PBT. We organise these factors at two levels: (1) the individual campaign around a candidate/party and (2) the national system (i.e., the electoral system, regulatory framework, and culture). This translates into the model shown in Figure 1, which will be elaborated on in the next paragraphs.

Figure 1: Factors influencing the adoption of PBT

The campaign team level

In his extensive research of US political campaigns, Daniel Kreiss (2016) identified four factors concerned with technological innovation within political campaigns. There are resource factors, such as campaign budgets and the number of volunteers a campaign can employ; infrastructural factors, such as technological tools or skills within the organization; organizational factors, such as organisational culture and structure; structural electoral cycle factors, such as election results. Building upon Kreiss' factors, we add an additional four (one campaign team level factor and three system level factors) to examine the use of PBT. On a campaign team level, the factor is ethical and legal concerns, such as normative reservations towards PBT. On a system level, the factors are electoral context, regulatory framework,and culture (discussed below). These new factors were identified through a review of literature about innovation in data-driven political campaigning techniques (e.g., Anstead, 2017; Kreiss, 2016; Jungherr, 2016; Hersh, 2015; Nielsen, 2012), and literature about (hybridisation of) campaign evolvement (e.g., Lijphart, 2012; Plasser & Plasser, 2002; Karlsen, 2010; Norris, 2000).

Resource factors

The main elements within this factor that could influence the extent to which campaigns can use PBT-techniques are: the budget and the effort needed to carry out a PBT-operation. A large budget enables campaigns to hire skilled personnel, acquire data, or buy targeted ads. The same dynamic applies to the number of volunteers a campaign can mobilise: having a lot of them facilitates a campaign in collecting data by canvassing, and sending potential voters targeted messages (the use of volunteers, of course, is dependent on their skills). Having a small budget and few volunteers, consequently, can be a barrier for campaigns because it bars them from acquiring the same amount of capabilities or from carrying out an operation on a large scale. This is in line with normalization theory (Margolis & Resnick, 2000), according to which the possibilities of the internet will not upset traditional power structures, but will rather develop along traditional lines as in the 'offline world'.

We can also view PBT as a means of using a campaign's resources as efficient as possible, to ensure parties do not spend money and effort on voters who will vote for another party anyway, or on citizens who will not vote altogether. Then, parties with limited resources could be more inclined to use PBT to not waste precious money, time, and labour. This is in accordance with the idea of equalization, which views the internet as an empowering tool for smaller parties due to its low costs and its new ways of direct communication with the electorate (Margolis, Resnick, & Levy, 2003; Bimber & Davis, 2003; Stanyer, 2010). A meta-analysis found evidence for the existence of both normalisation and equalisation in election campaigns (Strandberg, 2008). The occurrence of either process can differ per country and is dependent on several contextual factors, which will be discussed later on.

Organisational factors

The elements in this factor are about how campaign leaders perceive campaigning. Do they rely on proven best practices from previous campaigns or is there a culture of innovation? John Padgett and Walter Powell (2012) describe the concept of network folding. Applied to the political realm, this entails the extent to which campaigns employ skilled personnel from non-political sectors and to integrate that expertise into their existing institutions. An example is the hiring of Google engineer Stephanie Hannon as chief technology officer by the Clinton campaign (Easton, 2015). The 'cognitive diversity' following from network folding can lead to creative ideas (De Vaan, Stark, & Vedres, 2015). Furthermore, the organisational structure can be expected to resemble the way the campaign perceives PBT. A campaign with an autonomous data department is probably more prone to rolling out a PBT-strategy than a campaign that sees 'data' as only one of the many tasks of a communication staffer. Also, a change in leadership can be a facilitator for innovation (Gibson & Römmele, 2001).

Infrastructural factors

Elements are the technological tools available to campaigns, which enable them to roll out a PBT-operation. For instance, such tools might assist volunteers in the field by enabling them to collect data. They can be developed in-house or outsourced; in fact, there are specialised third party consultancies, that offer off-the-shelf tools, which in turn allow campaigns to employ innovative technology even though the campaigners do not have any technical expertise.

Structural electoral factors1

The actions of rival campaigns fall under the umbrella of structural electoral factors. A successful PBT-campaign of a rival can facilitate innovation in other campaigns, especially if those other campaigns themselves look back at an unsuccessful election. This connects with the 'critical event' (Kreiss, 2016), such as losing an election that should have been won, or with the experience of an 'external shock', which can be an incentive for professionalisation (Gibson & Römmele, 2001).

A second element influencing campaigns' likelihood to use PBT-techniques, is issue ownership (Petrocik, 1996), and the subsequent statements of party candidates propagating standpoints of the party. A political campaign 'caught' using privacy-infringing PBT-techniques, while its candidates present themselves as privacy champions, is likely to come across as hypocritical. Being perceived as such should be avoided, considering the negative electoral consequences following political-ideological hypocrisy perpetrated by politicians (Bhatti, Hansen, & Olsen, 2013).

Ethical and legal concerns

Elements within this factor consist of ethical and legal restrictions on how campaigns operate. For example, a political party could believe that PBT is ethically wrong as it infringes on citizens' right to privacy, and citizens' autonomy to form their own opinions. As a result, the party 'self-regulates' and refrains from using campaigning techniques violating its ethical beliefs.

Another element is the legal uncertainty that occurs when a campaign does not know how to behave in accordance with data protection and election laws, because of a lack of internal expertise. Such confusion can result in differences in the actions taken by comparable actors (e.g. Raskolnikov, 2017). Legal uncertainty can lead to 'overcompliance', which can be seen as a barrier towards the adoption of PBT-techniques, or to 'undercompliance', which facilitates the adoption of PBT-techniques (Calfee & Craswell, 1984). For instance, Anstead (2017) notes how parties felt disadvantaged by targeting possibilities facilitated by the perceived undercompliance with UK campaign finance law during the 2015 general elections.

The system level

Aside from campaign level factors, we look at contextual factors as well. These factors may limit the extent to which (US-American) campaigning techniques can be adopted in other countries (Karlsen, 2010). Therefore, we add three new contextual factors to our model. We expect that the electoral system, the regulatory system, and the culture of a democracy influence the extent to which the campaign team level factors are applicable. Below, we explore how the adoption of PBT-techniques can be influenced by properties of different systems. We will later apply our model (see Figure 1) to one specific case.

Electoral system

The three dominant electoral systems are first-past-the-post (FPTP), proportional representation (PR), and two-round (TR) (Birch, 2001, 2003). How these systems function, can influence how campaigns are run. The FPTP-system, first, can lead to an overvaluation of some key districts. Such districts sometimes 'swing' to one party and sometimes to another party, whereas other districts go to the same party in each election. As an effect, campaigns in a FPTP-system are inclined to spend a disproportionate amount of money and labour in these key districts in the hope of swinging the election their way (e.g., Anstead, 2017; Lipsitz, 2004). The PR-system, second, does not favour a select group of voters in a few key districts (Plasser & Plasser, 2002). This is especially true when the PR-system consists of only one district, in which every vote counts equally. As a result, campaigns have to spread their means more equally over the country. The TR-system, third, makes for a relatively unpredictable campaign, since it often is unclear which candidates will make it to the second round. Furthermore, the TR-system makes it important for campaigns to collect the votes of the supporters of the losing candidates of the first round. Therefore, campaigns should not only focus on their own base but other candidates' bases as well (Blais & Indridason, 2003). This has consequences for PBT, since campaigns should not only correctly classify potential voters as their own, but the other voters as well in order to target them in the next round.

A different aspect of electoral systems that influences how a campaign is run, is the degree of fractionalisation in a democracy (Duverger, 1959; Lijphart, 2012; Wang, 2012). FPTP-systems favour relatively few candidates/parties. PR-systems, in contrast, enable a large number of parties to run in an election. The first round of a TR-system can consist of many different candidates. As a result, campaigns that operate in a PR or a TR-system are less likely to launch attack campaigns against competitors. This is because PR-systems generally require a coalition of parties working together after the elections (Plasser & Plasser, 2002). And in a TR-system, campaigns should not mistreat competing candidates too much because winning campaigns have to court the bases of losing candidates in order to win the second round. Furthermore, in a highly fractionalised democracy, parties represent different (minority) groups within the electorate. This results in a high risk of 'mistargeting', in which campaigns approach a member of group A with appeals for a member of group B. Mistargeting can lead to voters penalising the campaign for their mistake (Hersh, 2013). These contextual circumstances may call for different PBT-strategies.

Regulatory framework

We distinguish strictly regulated campaign environments, moderately regulated campaign environments and minimally regulated campaign environments (Plasser & Plasser, 2002). Strictly regulated environments are characterised by "severe restrictions on the contact and communication between candidates and their constituencies" (p. 137). Moderately regulated environments typically focus on regulating access to TV-advertising and campaign funding. Minimally regulated environments impose few regulatory restrictions on political campaigns. It may be infeasible to implement innovative PBT-techniques in strictly regulated environments. Legal uncertainty can play a role on a system level too (e.g. because of a gap in the law).

Culture

Differences in the adoption and use of innovative PBT-practices can also be influenced by the culture or tradition in a democracy. For example, turnout culture is important because campaigns operating in countries where turnout is high will focus more on convincing voters on getting out the vote than campaigns operating in a low-turnout culture. In a low-turnout culture, campaigns sometimes target specific groups of voters (e.g. the elderly, who are more likely to turn out) more than other groups (e.g. the poor), who are unlikely to turn out (Herrnson, 2001). The turnout culture can influence the data campaigns collect on someone (and how campaigns tailor their messages), because a campaign message meant to convince someone typically leans on more data than a message meant to mobilise a voter does. Furthermore, cultural norms can dictate the strategy of political campaigns. In Japan, for example, posting dark post attack ads, such as the 'super predator' ad Trump launched against Clinton (Green & Issenberg, 2016), is improbable because of the cultural convention of averting direct conflict (Plasser & Plasser, 2002).

System level context is likely to affect campaign level factors. A campaign operating in a multiparty PR-system needs to pour more resources into identifying potential supporters than a campaign in a FPTP-system. After all, identifying potential Republicans or Democrats is easier than identifying potential voters in a ten-party race. Moreover, other than in US campaigns, most European campaigns are unable to access voter registration files provided by an electoral register. In many countries citizens can just show up at the voting booth, which means that the whole act of ‘registering’ for voting, as it is the case in the US, does not exist. Since Hersh (2015) has found that voter lists are one of the most valuable pieces of data to US campaigns, this principal unavailability, or non-existence, of such data poses a challenge for the PBT-capabilities of campaigns. But this challenge should by no means imply a PBT-operation in Europe is impossible. We would argue that while the lack of access to voter lists makes it very difficult to achieve the same level of granularity when engaging in PBT as in the US, by using other commercially available or self-collected data, it can be possible to come reasonably close to the desired level (which may be more modest for European campaign leaders than for American ones). The extent to which there is an actual difference between the degree of granularity between US and European campaigns, however, is outside of the scope of this paper, as we focus on perceptions and strategies of campaign leaders.

System level context also affects infrastructure. For instance, should the groundwork be spread equally across the nation, or focused on a number of battleground states? Also, a campaign operating in a heavily regulated context is likely to encounter some legal barriers. For example, because of campaign financing regulations (may influence resource factors), and data protection regulations (infrastructural factors). The absence of regulations, conversely, can facilitate PBT. Cultural context, finally, can influence campaigns' ethical considerations regarding PBT. Campaigns operating in a culture that favours privacy, for example, can be expected to avoid (or use less-invasive) PBT-techniques than campaigns run in a culture in which privacy is less important. In sum, there are several factors, both on campaign and system level, which can form a barrier or facilitate the extent to which campaigns are able to use PBT-techniques and how they use them.

Extending existing research to a European context, we have developed and will apply an improved model (applicable in different electoral contexts) to analyse barriers and facilitators to innovative PBT-practices by political campaigns. As the context of the research case differs from the US, we expect to contribute to the framework and to shed light on how contextual factors influence innovation of political campaigns. Furthermore, in answering our research question, we provide insight into the way political campaigns in a multiparty democracy organise, communicate and innovate. Given these considerations, our key question is: What barriers and facilitators for the adoption and use of PBT-techniques do Dutch political parties perceive?

Method

This study focuses on campaigns in the Netherlands because of the national elections taking place in the research period (15 March 2017), the advanced technological infrastructure (Coy, 2015), and the interesting contextual factors. The Dutch electoral system is one of open list proportional representation (PR), in which all members of parliament come from one nationwide district (Lijphart, 2012). This means that in the Netherlands, every vote counts equally. Moreover, the system of PR (and the very low de-facto threshold) enables a relatively large number of political parties to run in an election. 28 parties participated in the 2017 national election ("Partijen nemen deel", 2017). Of these parties, 13 actually gained a seat in parliament ("Officiële uitslag", 2017).

The Dutch national elections have a relatively high turnout: around 80% in the previous two elections ("Officiële uitslag", 2017). But where the US presidential campaigns can spend hundreds of millions of dollars (Narayanswamy, Cameron, & Gold, 2017), the Dutch campaign with the biggest funds (VVD) has no more than 5 million dollars to spend. And even if the budgets were sufficiently large, the question is whether voter data would be usable for a political campaign. The Dutch data protection law categorises political preference as sensitive personal data. This means that campaigns are only allowed to process such information if the potential voter explicitly gives permission to do so.

Finally, as party membership in the Netherlands steadily decreases, political campaigns can rely less on their members to do labour-intensive tasks (such as canvassing). In 2016, the number of party members of all political parties combined, was at its lowest point since the second world war. Although this number has picked up slightly since, party membership is still quite low ("Membership Dutch parties still low", 2017).

After approval from the ethical committee of the University of Amsterdam, we carried out eight in-depth interviews with campaign leaders. We interviewed 11 campaign leaders, belonging to eight political parties in total (three interviews were double-interviews). In addition to this, we held two background interviews (with one local campaigner for the municipality of Amsterdam and one political consultant offering PBT-services). The eight elite-interviews on average lasted 53 minutes. Two were conducted by phone, the others face-to-face. We took a qualitative research approach for several reasons: the small group of people concerned with the coordination of political campaigns in the Netherlands, the lack of knowledge on this topic in the Netherlands, and because interviewing is a suitable method for understanding the mechanisms behind and perceptions of a phenomenon (Boeije, 2005). As we want to understand how campaigns see PBT, what they are actually doing, and how they perceive possible barriers and facilitators to the adoption and use of PBT-techniques, the interview is a suitable data-collection method. Using an interview guide (see appendix A), we held semi-structured interviews, allowing follow-up questions.

Interviewees

We selected the interviewees via purposive sampling. Campaign leaders qualified for an interview when they had a coordinating role in the campaign and were campaigning for a party that gained at least one seat in the 2012 national parliamentary elections. Eleven campaigns satisfied this second criterion (see Table 1). We contacted interviewees via email, explaining the objective of the study. The interviewees signed an informed consent document before the interview started. We also promised the campaign leaders anonymity, and confidentiality until after election day (15 March 2017). By doing so, we tried to provide the interviewees with a safe environment in which they felt like speaking freely, and without concern of somehow 'leaking' strategic information. Because no information would become public before election day, the risk that interviewees might provide biased information due to a strategic agenda was minimised. Another large advantage of interviewing the campaign leaders before the elections took place, is the prevention of hindsight bias by the interviewees. Unfortunately, we were unable to convince three parties to comply (VVD [right on the political spectrum], PVV [right-wing nationalist party], PvdD [Party for the animals; left-wing]). These three parties were unwilling to cooperate, either because they still found the risk of leaking their strategy too large, or they did not offer an explanation.

Analysis

Using a grounded theory approach, this study has passed four phases: the exploration phase, the specification phase, the reduction phase, and the integration phase (Wester, 1995). In the exploration phase, two background interviews took place (with a campaigner for the municipality of Amsterdam and with a political consultant offering PBT-services). These were coded using AtlasTI, 'tentatively labelling' relevant information (Glaser, 1978). Thereafter, the first interviews with campaign leaders took place. These were transcribed and open-coded. Furthermore, fellow researchers also coded these interviews and discussed the content (peer debriefing). In the next phase, new interviews took place and the data was subject to axial coding. The first dimensions were identified (e.g., what forms a barrier and what facilitates the use of PBT-techniques?). The reduction phase saw the emergence of the core category (innovation). In the integration phase, we completed the conceptual framework, finalised our analysis, and had the campaign leaders approve the quotes used (member checking). This means the campaign leaders agreed with the way they were quoted, and with the publishing of the names of the political parties. Member checking increased the willingness of campaign leaders to cooperate with the study.

Results

We first describe the field: to what extent do campaigns use PBT-techniques? Then we explain differences between parties by focusing on the five campaign level factors concerning the use of PBT (resource, infrastructural, organisational, structural electoral cycle factors, and ethical and legal concerns). Finally, we zoom out to the system level and discuss the influence of contextual factors (electoral system, regulatory framework and culture) on the adoption of PBT-techniques.

PBT in Dutch campaigns

As campaigns in the Netherlands can have recourse to relatively detailed public census data and detailed election results, all campaigns, to some extent, adopt a PBT-approach. Furthermore, Facebook is an important tool for all parties, but the parties differ in how they use Facebook's capabilities. Some parties occasionally post content targeted to broad age groups, while other campaigns frequently post content tailored to more specific groups. Two campaigns stand out, as they have developed their own PBT-tools, which they can use to continuously refine their knowledge of the electorate. We will now use our model to explain the differences between campaigns.

Resource factors

All campaigns cite financial costs as a barrier. Table 2 shows that budgets are modest, and differ between parties.

Source: parties' annual financial reports, on file with authors.

These small budgets form a barrier for the cooperation with expert political consultants (such as Blue State Digital) to enhance their PBT-operations. Parties refer to the financial costs as the main reason not to hire consultants. Campaign leader 1 of the Social Democratic Party (PvdA) explains why he does not work with Blue State Digital (BSD):

Their system is very expensive, that's a factor. And you need the people to carry out the work for you. In an ideal world, such a cooperation would be really cool though."

Liberal democrat party D66 agrees: "because it costs a lot of money and we don't have that kind of money. And if we spend it on a consultant, we can't spend it on the campaign itself."

The same barrier appears when campaigns speak about other technological means, such as canvassing apps, allowing campaigns to directly process information from canvassers. Christian Democrats CDA, for instance, would like such an app. Campaign leader 8: "yes, but that would demand a financial investment that we can't afford."Green Party GroenLinks has a contrasting perspective: "I believe it usually costs around €100,000 to build an app such as our own. (..) We, however, paid our programmers two crates of beer and 40 pizzas." Several facilitators help GroenLinks and also socialist party SP to overcome this barrier of financial costs. First: the personal network of the campaign leader. This facilitator is especially prominent for GroenLinks, where campaign leader 7 employs his own network to optimise the BSD-systems, but also to help him with setting up other parts of the campaign:

We had to adjust it [the BSD system] somewhat for the Netherlands. The people with whom I did so, Swedish folks ... they are simply a little network of people of around my own age, and some people who are a bit older and have already set up a similar campaign in their own countries. A guy who set up the grassroots organization for Trudeau, for example, he's a couple of years older than I am, but I Skype with him to talk about how I should handle certain things.

Campaign leader 7's personal network plays (or at least played)an important role in cheaply setting up technological tools and creating content:

Through the network, I'm aware of the crowdfunding streams for a normal campaign. (..) I'm meeting a friend tomorrow, who has experience with mail flows. (..) I have a network of volunteering writers, poets, freelance journalists who write for us for free. (..) So partly, I just have a good personal network.

However, as PvdA notes, having lots of data is of no use if you don't have the capacity to use it. GroenLinks tries to overcome this barrier by organising their campaigns, to some extent, in a citizen-initiated manner (Gibson, 2015). A citizen-initiated campaign (CIC), devolves "power over core tasks to the grassroots" (p.183). As campaign leader 7 puts it:

Grassroots is about creating an infrastructure to enable as many sympathizers as possible to volunteer as canvassers on a large scale. So voter contact on a large scale, but also – and that's Bernie's [Sanders] lesson – to have places in which a few people make stuff by themselves without us having any control over it. (..) Embracing people's creativity without managing it.

SP has less need of a citizen initiated campaign, because of their relatively large number of active party members. "The big difference [with GroenLinks] is the fact that we already have the volunteers. Many other parties lack the numbers. We have thousands of party members who gladly canvass for us two weekdays and on Saturday as well." [Campaign leader 11, SP].

Infrastructural factors

Having a good infrastructure allows campaigns to actually collect data and send tailored messages. What kind of PBT-infrastructure can parties rely on and how does it facilitate their use of PBT?

All campaigns use the PBT-infrastructure Facebook offers, although some more than others. Nearly all campaigns use its lookalike audiences function to find new potential voters. Campaign leader 8: "we search for profiles of people who look like the ones who’ve already liked our Facebook page, and then serve them with advertisements." Campaigns also look at people who like pages that are close to the values of the political parties. Christian party ChristenUnie, for instance, tries to target voters who like the page of evangelical broadcaster EO. So does the Calvinist Political Party SGP, which tries to find out people’s interests on Facebook: "For example… farming, or Israel, off the top of my head; you try to approach people along the lines of their interest, or the region in which they reside." [Campaign leader 9]

Some campaigns also employ ‘dark posts’, a Facebook function that enables campaigns to opaquely target specific audiences, while its messages are not visible to untargeted Facebook users. Campaign leader 1 exemplifies:

We’ve managed to get something done related to gas extraction in Groningen. It doesn’t make sense to share that on the national Facebook page, because it was only important news locally. So we put out a dark post, only for Groningen residents. Sometimes we can specify it even more.”

Using Facebook for PBT-purposes, campaigns do not actually gather or own data themselves. There are a few campaigns that do gather their own data, by using canvassing apps. Campaign leader 7:

We use the election results per voting location and use that information to establish the GroenLinks mindedness of a neighbourhood. Then we can prioritise which addresses to visit and which to ignore. When we visit addresses, our volunteers use the app to answer the following questions: 1. Is anyone home? 2. Does she want to talk? 3. Is she going to vote? 4. Is she planning to vote for GroenLinks? 5. What is the most important theme to her? 6. How GroenLinks minded was she? If she considers to vote for GroenLinks, two questions follow: 1. Do you want to stay informed of our campaign by e-mail? 2. Can I have your phone number, so we can ask you to do canvassing talks?”

The GroenLinks app facilitates large scale collection of information about people's political preferences, thereby informing strategic decisions. Also, the personal data can facilitate accurate PBT on an individual level. The secondary objective of the app is to provide an infrastructure for volunteers to campaign on their own terms, whenever they feel like doing so:

Our app, built by hackers, enables others to campaign for us. (..) Someone in [small town] Lutjebroek can install our app and go ahead and work for our campaign. No campaign leader needed. [Campaign leader 7]

Some campaigns monitor the visitors of their own websites. Campaign leader 1: “What are people searching for on our website, how do they get to our website, how much time do they spend, (..) which button should you colour red? How does that work?” At the time of the interview, CDA was not yet tracking their website visitors, but: “we’ve just migrated to a new website, on which we want to start collecting more data on our visitors. I’m curious what kinds of people are visiting the website. And what kinds of people don’t, and therefore have to be reached through different channels.”

SP has built a system which combines previous election results, census data and their own membership Constituent Relationship Management (CRM) data. Plotted on a Google Map, they can identify interesting areas for them to canvass. This system facilitates efficient use of means:

We would do nothing more happily than knocking on every single door in every city, but unfortunately, we do not yet have that kind of manpower. So we do an analysis: What kinds of neighbourhoods are especially interesting for us? We have built our own system to help us make that decision [Campaign leader 11].

Organisational factors

Circumstances within the campaign’s organisation itself can form a barrier for the uptake of PBT-techniques. Less innovative parties, for instance, do not have a dedicated data, tech, or digital department. As a campaign leader notes: "The department responsible for that [tech/data/digital] is our Communication department. So that's four or five people. And sometimes someone of the department picks it up, but there's not one specific person who's responsible."This contrasts with GroenLinks, which has a Digital and Grassroots department and with SP's Digital department.

The "state of mind"within a campaign can also be seen as a barrier:"In the sense that internally, people are still very much inclined to think offline. The culture within the campaign is quite offline." [anonymous campaign leader]2

New leadership and younger staffers can play a facilitating role in political organisations. Campaign leader 11 argues that, because he is young, their new party chair brings a more tech-savvy vision than his predecessor. According to the campaign leader, younger staffers are more likely to implement tech and data in their work procedures.

A final organisational barrier is the primary goal a political party pursues. Campaign leader 10:

Maybe the strange thing about SGP is that we do not care that much about seat maximization. For us, it's about the impact of our principles. And sure, we would rather have four seats than three, but if we have to settle for three seats: that's fine too. And that's, in my opinion, a reason why we have a feeling like: do we really need data?

Structural electoral cycle factors

These circumstances are largely beyond the control of the campaigns, but they can influence the uptake of PBT-techniques. Campaign leaders see the PBT-actions of other political campaigns as a motivational factor. As campaign leader 11 notes about the development of their app: "I've looked a little bit at how GroenLinks have their app and canvassing system." Or as campaign leader 2 concludes: "If every party does it, you don't win very much by it. But if you're the only party that does nothing..."

Ethical and legal concerns

Especially D66 and the seniors’ party 50PLUS take a principled stance against the collection of data and the use of PBT. Where D66 presents itself as a privacy champion and therefore will never gather and use information about (groups of) voters, 50PLUS campaign leader 6 warns about the risk of irresponsible use of the data gathered by the "almost stalking of people", which he calls "morally irresponsible".

Furthermore, a lack of internal legal expertise appears to contribute to a feeling of legal uncertainty, which affects the likelihood of adopting PBT-techniques: "Legislation has grown so very comprehensive and complex. It's almost impossible to cope for us as a small organisation." [Campaign leader 5]

While ethical and legal concerns can form a barrier, a left- or right-wing orientation does not seem to be instrumental therein. After all, we have seen left-wing parties GroenLinks and SP develop relatively advanced PBT-tools. And we have seen right-of-centre party CDA express clear interest in advancing their own PBT capabilities. At the same time, left-of-centre 50PLUS and right-of-centre D66 both oppose the use of PBT.

System level

Electoral system

Although the Dutch one district PR-system should make for a rather equal distribution of campaign efforts, campaigns still divide the country into smaller areas of interest called 'key areas'. These areas differ per party, but do receive a relatively large part of campaign attention. Campaign leader 1 describes these as areas: "where we know the turnout is low, but the number of PvdA-voters is high". All campaigns use data provided by the Electoral Council, showing the election results per party, per voting location to establish key areas. Campaign leader 3 explains:

Using that [the election results], you see: Okay, we do well in this neighbourhood or this street. And then you combine that information with the CBS3 data, to find out what kind of neighbourhood it is, what kind of people live there, what are their backgrounds, how much do they earn, what does the family composition look like, et cetera.

Facilitated by these public data, campaigns enrich their knowledge of specific areas. A next step would be to use those data to make personalised appeals to (subgroups of) people living in those specific key areas.

Regulatory framework

Although the Netherlands would qualify as a minimally regulated environment (Plasser & Plasser, 2002; Esser & Strömbäck, 2012), campaigns all experience regulatory pressure and legal uncertainty on a system level. They cite an abundance of regulations, forming a barrier to their ability to innovate.

The technological developments have been taking place so very quickly. And, in that timeframe, to adjust all your procedures and everything. And also to meet the privacy regulations, I think many parties face a huge challenge in that respect." [Campaign leader 4]

Campaigns sometimes face a dilemma, having to decide between innovative techniques and privacy regulations. Campaign leader 11:

Regulations sometimes are unclear, which leads us to decide to go for the safe option because you do not know where the red line is. And you never want to abuse someone's personal data. So yes, regulations sometimes cause us to hit the brake and that's a good thing.

Culture

There is a recurring worry about the perceived low level of political knowledge of the average Dutch voter. PBT-techniques can facilitate campaigns' efforts to convince or educate such low-information voters, for example by "having a conversation with someone, especially if you share some characteristics," [Campaign leader 7] or by interesting "people for things that are relevant to them and to make them aware of the political dimension of those things." [Campaign leader 8] Campaign leader 3, in contrast, concludes that the electorate's low level of political knowledge (together with the perceived volatility of the electorate, the decrease in political trust, and their focus on persons instead of parties) forms an insurmountable barrier, making PBT-techniques irrelevant.

Discussion and conclusion

In the 2017 elections, used here as a case study, all campaigns use PBT through Facebook, but some parties are more advanced than others, and have even developed their own PBT-tools. We have established what the main barriers and facilitators for PBT are, using five factors on a campaign level and three factors on a system level. Not only does this study shed light on the conditions under which these barriers and facilitators manifest themselves, it also gives insight into their different workings across parties. Our study provides information about the data collected by parties and the PBT-techniques used to attract voters. We demonstrate how personal networks and cognitive diversity within a campaign can level barriers. We show how PBT is not only perceived as useful for campaigns in a FPTP-system, but in a PR-system as well. And we show how regulatory pressure is perceived as an obstacle and as a welcome 'normative red line'.

A triangulated research approach can improve our understanding of the campaign leaders' constructs. Observation of their (use of) PBT-tools and how these tools help campaigns make strategic decisions, can give more insight into the workings of these techniques. Another approach would be to interview canvassers and identify 'field-level' barriers and facilitators. Furthermore, ideally, we would have spoken to all parties holding a seat in parliament. Unfortunately, three parties did not cooperate. Two of those became the largest (VVD) and second largest (PVV; in a very close field) party. Since we did have access to eight of 11 parties, we are confident about our findings and we do not expect to identify additional factors influencing the adoption of PBT from interviews with the remaining parties.

Compared to related recent studies by Anstead (2017), Hersh (2015), Kreiss (2016), and Nielsen (2012), this study makes a number of contributions. In general, we focus our exploratory research on a PR-system instead of a FPTP-system, and we develop a model that takes system level contextual factors into account. Specifically, unlike Anstead (2017) we have found evidence for equalisation (which occurs when smaller parties take advantage of the internet's low costs and direct communication possibilities, and, in doing so, use the internet as a tool of empowerment [e.g. Margolis, Resnick, & Levy, 2003]). This evidence is especially clear in the case of GroenLinks, which was, at the time of the campaign, one of the smallest parties in parliament (now the fifth party). Furthermore, we provide an insightful point of view into Anstead's question of whether "parties develop data-driven capabilities more rapidly in electoral systems with a tendency towards disproportionate outcomes" (2017; p. 23). In comparison with Hersh (2015), we focus less on how differences in data-availability lead to different strategic decisions, and more on how differences in the perception of campaign level and system level factors lead to a variation in the occurrence in PBT-innovation. With regard to Kreiss (2016), we have extended his model and applied it to a multiparty democracy. In comparison with Nielsen (2012), we focus solely on the perception of campaign leaders and not on canvassers. Furthermore, we focus on PBT on online as well as offline platforms.

Our attention for the system level factors has enabled us to identify perceived influence of the PR-system on the adoption of PBT. Contrary to theoretical expectations (Plasser & Plasser, 2002), campaigns in a one-district PR-system do identify key-areas that are more heavily campaigned than other districts. These key-areas differ from 'battleground states' in FPTP-systems in the sense that the key-area does not sometimes swing one way and sometimes the other, but rather that potential voters in key-areas are supportive of a certain party, but not very likely to show up at the polls. Campaigns use PBT-techniques to convince these potential voters of the personal relevance of politics and to motivate them to cast their vote. Areas with firm turnout numbers and clear support for a certain party, in contrast, are perceived as less decisive and less of a priority. This leads to a hierarchy of areas, which differs per party. Also, as a PR-system typically leads to a relatively large number of parties partaking in an election, PBT can be seen as an asset for a campaign to organise in a more efficient manner. Moreover, according to the campaign leaders, PBT-techniques offered by Facebook do allow smaller parties a degree of visibility that they are unable to achieve through traditional media.

On a campaign level, in the coming years, we expect more citizen-initiated campaigning (Gibson, 2015) by campaigns low in labour-resources. This requires a solid infrastructure, which opens the door for third party intermediaries offering off-the-shelf infrastructure. In this regard, it would be interesting to track the development of PvdA, which has suffered its biggest loss in history. This critical event could lead to the prototyping (Kreiss, 2016) of GroenLinks' innovative campaign by PvdA. As the party's chairman has resigned, the door is open to a more cognitive diverse party structure (Du Pre, 2017; De Vaan et al., 2015). Of course, these developments might apply less to parties that are officially more cognisant of campaign ethics (e.g. D66). This is why ethics and legal aspects are important factors to take into consideration. It would be interesting to see how these campaigns act as PBT-capabilities of rival parties improve. Their self-imposed barrier can limit their future chances, but can also attract voters growing more aware of the value of privacy. In the former case, this could lead to an overhaul of their privacy principles, or perhaps to a legislative push towards the restriction of PBT (similar to Hersh, 2015). In the latter case, campaigns can be expected to develop innovative non privacy-invasive campaigning techniques. Either way, our model would provide tools to study the process.

So 'can political parties campaign in Europe as they do in North America' (Bennett, 2016)? We would say ‘mostly yes’. We agree with Bennett (2015) that there are important differences between the US and Europe, and indeed, they influence how PBT is used. But based on our findings, we are hesitant to conclude that those differences (severely) constrain the export of PBT-practices to European multiparty systems. We have shown that relatively small campaign budgets do not need to bar parties from engaging in PBT-practices (or even from cooperating with BSD, an 'expensive' American political consultancy). The same is true of the electoral system: campaign leaders generally perceive PBT-techniques as useful in a PR-system. What remains is the relatively strict Dutch data protection law, labelling political preference as 'sensitive personal data', which can only be processed with explicit consent from the potential voter. 'Explicit consent', however, sounds harsher on paper than it is in practice and is easily achieved (e.g., Beales & Muris, 2008; Calo, 2012; Joergensen, 2014). Of course, because of data regulations and/or their non-existence, European campaigns are unable to consult voting lists showing whether an individual showed up at the polls in the last elections. In most European countries, the electoral register is inaccessible to political parties. One might argue that, from a campaign's perspective, US voter data are superior to European voter data. We would argue that European data are different, but they do not bar European campaigns in the use of PBT-techniques. Dutch campaigns, for instance, can (and do) rely on election results on voting booth level (which comprises a couple of streets). They can (and do) combine these results with detailed, accurate, and a multitude of data about the neighbourhoods surrounding those voting booths. And then there is Facebook, facilitating easy targeting of its users with personalised messages. As potential challenges for democracy come with PBT, such as ignoring 'less valuable' citizens (e.g. reliable non-voters), more research into the workings and effects of PBT is needed.

Disclaimer: This guest essay in the Special issue on political micro-targeting has not been peer reviewed. It is an abbreviated version of a speech delivered by the Member of the European Partiament (MEP) Sophie in ‘t Veld in Amsterdam in May 2017 to Data & Democracy, a conference on political micro-targeting.

Democracy

Democracy is valuable and vulnerable, which is reason enough to remain alert for new developments that can undermine her. In recent months, we have seen enough examples of the growing impact of personal data in campaigns and elections. It is important and urgent for us to publicly debate this development. It is easy to see why we should take action against extremist propaganda of hatemongers aiming to recruit young people for violent acts. But we euphemistically speak of 'fake news' when lies, 'half-truths’, conspiracy theories, and sedition creepily poison public opinion.

The literal meaning of democracy is 'the power of the people'. 'Power' presupposes freedom. Freedom to choose and to decide. Freedom from coercion and pressure. Freedom from manipulation. 'Power' also presupposes knowledge. Knowledge of all facts, aspects, and options. And knowing how to balance them against each other. When freedom and knowledge are restricted, there can be no power.

In a democracy, every individual choice influences society as a whole. Therefore, the common interest is served with everyone's ability to make their choices in complete freedom, and with complete knowledge.

The interests of parties and political candidates who compete for citizen’s votes may differ from that higher interest. They want citizens to see their political advertising, and only theirs, not that of their competitors. Not only do parties and candidates compete for the voter's favour. They contend for his exclusive time and attention as well.

Political targeting

No laws dictate what kind of information a voter should rely on to be able to make the right consideration. For lamb chops, toothpaste, mortgages or cars, for example, it’s mandatory for producers to mention the origin and properties. This enables consumers to make a responsible decision. Providing false information is illegal. All ingredients, properties, and risks have to be mentioned on the label.

Political communication, however, is protected by freedom of speech. Political parties are allowed to use all kinds of sales tricks.

And, of course, campaigns do their utmost and continuously test the limits of the socially acceptable.

Nothing new, so far. There is no holding back in getting the voters to cast their vote on your party or your candidate. From temptation with attractive promises, to outright bribery. From applying pressure to straightforward intimidation.

Important therein is how and where you can reach the voter. In the old days it was easy: Catholics were told on Sundays in church that they had no other choice in the voting booth than the catholic choice. And no righteous Catholic dared to think about voting differently. At home, the father told the mother how to vote. The children received their political preference from home and from school. Catholics learned about current affairs via a catholic newspaper, and through the catholic radio broadcaster. In the Dutch society, which consisted of a few of such pillars, one was only offered the opinions of one's own pillar1. A kind of filter bubble avant la lettre.

Political micro-targeting

Nowadays, political parties have a different approach. With new technologies, the sky is the limit.

Increasingly advanced techniques allow the mapping of voter preferences, activities, and connections. Using endless amounts of personal data, any individual on earth can be reconstructed in detail. Not only can their personal beliefs be distilled from large troves of data, no, it even is possible to predict a person's beliefs, even before they have formed them themselves. And, subsequently, it is possible to subtly steer those beliefs, while leaving the person thinking they made their decision all by themselves.

As often is the case, the Americans lead in the use of new techniques. While we Europeans, touchingly old-fashioned knock on doors and hand out flyers at Saturday's market, the American employ the latest technology to identify, approach, and influence voters.

Of course, trying to find out where voters can be reached and how they can be influenced is no novelty. Political parties map which neighbourhoods predominantly vote for them, which neighbourhoods have potential, and in which neighbourhoods campaigning would be a wasted effort. Parties work with detailed profiles and target audiences, for which they can tailor their messages.

But the usage of personal data on a large scale has a lot more to offer. Obviously, this is a big opportunity for political parties, and for anyone else, who runs campaigns or aims to influence the elections.

However, the influencing techniques become increasingly opaque. As a result of the alleged filter bubble, voters are being reaffirmed in their own beliefs, and they hardly receive information anymore about the beliefs and arguments of other groups. This new kind of segmentation may stifle critical thinking. There may not be enough incentive to test one's own ideas, to find new arguments, or to critically reflect on the truthfulness of information.

I am a social and economic liberal D66 politician, and I get suggestions for news articles from websites like The Guardian or Le Monde. My colleague from the right wing nationalist PVV, may well receive URLs from Breitbart.

Pluralism is essential for a healthy, robust democracy. In a polarised society, people live in tightly knit groups, which hardly communicate with each other. In a pluralist society people engage in the free exchange, confrontation, and fusion of ideas.

The concept pluralism is under pressure. Populist parties declare themselves representative of The People. In their vision, The People, is uniform and homogenous. There is a dominant cultural norm, dictated from the top-down, to which everyone must conform. Whomever refuses, gets chewed out. Often, it is about one-dimensional symbolism such as Easter eggs and Christmas trees. There is no place for pluralism in the world of the populists. But when there is no pluralism, there is no democracy. Without pluralism, democracy is nothing more than a simple tribal dispute, instead of the expression of the will of all citizens together.

Voter data

European privacy legislation limits the use of personal data. In the world of ‘big data’, one of the explicit goals of regulation is to prevent restriction of the consumer's choice. Oddly enough, lawmakers do not explicitly aspire to guarantee voters as broad a choice as possible. But in politics, individual choices have consequences for society as a whole.

In 2018, the General Data Protection Regulation (GDPR) comes into effect. We have worked five years on the GDPR. At this moment, we work on the modernisation of the e-Privacy Directive, which is mainly about the protection of communication. As was the case with the GDPR, companies from certain sectors scream bloody murder. European privacy protection would mean certain death for the European industry. According to some corporate Cassandras, entire European industries will move to other continents. That very same death of corporate Europe is also predicted for any measure concerning, say, environmental norms, procurement rules, or employee rights. All those measures are in place, but, as far as I know, the nightmare scenario has never occurred...

There are some corporate sectors, such as publishing and marketing, which have a huge impact on the information supply to citizens. They are the ones who now cry wolf. It is understandable that they are unhappy with stricter rules concerning their activities, but as the potential impact of the use of personal data and ‘big data’ increases, so does their social responsibility.

At the moment, there is not much public debate about the new techniques. Peculiar. Thirty years ago, 'subliminal advertising', as we called it then, was prohibited because people found it unethical to influence people without their knowledge. We need to have a similar debate. What do we think of opaque influencing? Do we need ethical norms? Should such norms apply only to political campaigns, or should we look at this from a broader perspective? In the ‘big data’ debate, we tend to speak in technical or legal terms, while actually the issue is fundamentally ethical, holding far-reaching consequences for the vitality of our democracy.

Such a public debate demands more clarity on the impact of ‘big data’, profiling, targeting, and similar techniques on the individual, her behaviour, and her choices, which determine in what direction society progresses. Which voters are being reached? How sensitive are they for the subtle influencing and what makes them resilient? How do people who are hardly reached only compare to the others? How do voters and non-voters compare? Is the voter truly predictable? Can we identify or influence the floating voter? Do voters actually float between different parties? Or do they especially float within their own party, their own bubble, their own segment? How important are other factors, such as the social context? If the new influencing techniques are indeed as potent as we think, how can polls get it so wrong? What can we learn from advertisers who return to contextual advertising, because targeting turns out less effective than they thought?

We need to stay cool-headed. New technologies have a huge impact, but human nature will not suddenly change due to ‘big data’ and its use. Our natural instincts and reflexes will definitely not evolve in a few years. That would take many thousands of years, as even in the 21st century, we seem to have more than a few cavemen traits, so losing internalised behaviour is not as easy as 1-2-3. Humans are resilient, but democracy is vulnerable. On a short term, the societal impact is large. This gives us all the reason to reflect on how to deal with the new reality, and how we can keep up our values in this new reality.

The use of personal data, clearly, is not solely reserved for decent political parties. Other persons and organisations, from the Kremlin to Breitbart, can bombard European voters with information and misinformation. But European governments, controlling endless amounts of personal data of their citizens, can also manipulate information, or circulate utter nonsense to advance their own interests. A random example: the Hungarian government influencing their voters with lies and manipulation about the so-called consultation on asylum seekers.

Beyond voter data

This issue is not only about the personal data of voters, but also about the personal data of political competitors, opponents, and critics, which are increasingly being employed. Recently, we have seen efforts of external parties to influence the results of the 2017 French elections. We saw a large-scale hack of the Emmanuel Macron campaign, and the spread of false information, coming obviously from the Kremlin and the American Alt-Right, meant to discredit Macron's candidacy.

Also, the American elections show the shady game of hacking, leaking, and manipulating. The issue of the Hillary Clinton mails will undoubtedly occupy our minds for years. Who knows how the elections would have turned out without this affair?

Other democratic pillars can get corrupted as well by the misuse of data. Critical voices, opposition, and checks and balances are democracy's oxygen. Democracy is in acute jeopardy when data are employed to attack, undermine, discredit, blackmail, or persecute journalists, judges, lawyers, NGOs, whistleblowers, and opposition parties.

In Europe, we tend to shrug our shoulders at these dangers. "Oh well, we'll see, such things occur only in banana republics, not right here". Of course, this trust in our democratic rule of law is wonderful. But if we treat our rule of law this neglectfully, we will lose it eventually.

Within the European Union, we currently see this happening in Poland and Hungary. The governments of both nations ruthlessly attack independent judges, critical media, inconvenient NGOs. They do so with quasi-lawful means. Under the banner of transparency, they force NGOs to register. In doing so, they misuse laws against money laundering, and terror finance. Or the governments bring out compromising information about judges or politicians in strategic moments.

But critical voices struggle in other member states as well. Lawyers are being monitored, even without a legal basis. In the years after 9/11, we have created endless new abilities for intelligence services, police and justice departments to spy on citizens, even without suspicion, without the signature of a judge. The companies to which we unwittingly surrender our personal data, in exchange for service, are forced to hand over all information to the government, or forced to build in backdoors. Governments hack computers in other countries. Usually, it starts out with unlawful practices, but soon enough laws are put in place to legalise those practices. The magic word 'terrorism' silences any critique on such legislation.

But when politicians, journalists, NGOs, whistleblowers, lawyers, and many others cannot perform their tasks freely and without worry, our democracy withers. Not only do they have to operate without someone keeping an eye on them, they have to know nobody is in fact watching them. The mere possibility of being watched, results in a chilling effect.

For this principal reason, I have contested a French mass surveillance law before the French Conseil d'Etat. Since, as a member of the European Parliament, I spend four days a month on French soil (in Strasbourg), I could potentially be the target of the French eavesdropping programme. This is not totally imaginary, as I am not only a politician, but also a vocal critic of certain French anti-terror measures. It is not about me actually worrying about being spied on, but about the fact that I might be spied on. Luckily, I am not easily startled, but I can imagine that many politicians are vulnerable. That is a risk for democracy.

I do not discard the possibility of a ruling of the European Court of Human Rights on my case. In that turn of events, it will lead to jurisprudence valid in the entire EU (and the geographical area covered by the Council of Europe).

But, of course, this should not depend on the actions of one obstinate individual whether politicians, NGOs, journalists, and so on, can do their jobs fearlessly, to fulfil their watchdog role.

It is my personal, deep, conviction that the biggest threat to our democracy is the fact that we have enabled the powerful to access, with almost no limitations, the personal data of those who should control those very same powerful entities.

What can we do?

Some propose new forms of democracy, in which universal suffrage is weakened or even abolished. In his book ‘Against elections: the case for democracy’, David Van Reybrouck had the idea to appoint representatives on the basis of chance, and in his book ‘Against democracy’ Jason Brennan wants to give the elite more votes than the lower classes, presuming that people with more education or development make better choices. Others want to replace representative democracy with direct democracy.

I oppose those ideas. Universal suffrage and the representative democracy are great achievements, which have led to enormous progress in society.

First of all, we have to make sure our children grow up to be critical, independent thinkers. Think differently, deviate, provoke: this must be encouraged instead of condemned. A democracy needs non-conformists.

We must teach our children to contextualise information and to compare sources.

The counterpart of ‘big data’ must be ‘big transparency’. We need to understand not just open administration, but also insights into the techniques of influence.

The regulation and limitation of the use of personal data, as I hope to have argued effectively, is not a game of out-of-touch privacy activists. It is essential for democracy. We need safeguards, not only to be sure people really are free in their choices, but also to protect the necessary checks and balances. As such, I plea for a rigorous application of the GDPR, and in the European Parliament, I will work for a firm e-Privacy Directive.

And yes, perhaps we should examine whether the rules for political campaigning are still up-to-date. In most countries, those rules cover a cap on campaign expenditures, a prohibition of campaigning or polling on the day before election day, or a ban on publishing information that may influence the election results, such as the leaked e-mails in France. But these rules have little impact on the use of personal data to subtly influence elections.

Last year, the European Parliament supported my proposal for a mechanism to guard democracy, the rule of law, and fundamental rights in Europe.2

On this day (editor’s note: 9 May, Europe Day) of European democracy, I plead for equal, high norms in Europe. The last years have shown that national elections are European elections. It is crucial for us to trust that all elections in EU member states are open, free, and honest elections, free of improper influencing.

These last sixty years, the European Union has developed itself into a world leader in democracy and freedom. If we start a public debate, Europe can remain a world leader.

This paper is part of 'A Manchurian candidate or just a dark horse? Towards the next generation of political micro-targeting research’, a Special issue of the Internet Policy Review.

Introduction

Political campaigns in the United States have employed digital technologies for more than a decade, developing increasingly sophisticated tools and techniques during each election cycle, as “computational politics” have become standard operating procedure (Tufecki, 2014; Kreiss, 2016). However, the most recent election marked a critical turning point, as candidates, political action committees, and other interest groups were able to take advantage of significant breakthroughs in data-driven marketing techniques, such as cross-device targeting, developed since the previous presidential election (“Bernie Sanders”, 2016; Edelman Digital, 2016). Electoral politics has now become fully integrated into a growing, global commercial digital media and marketing ecosystem that has already transformed how corporations market their products and influence consumers (Chahal, 2013; LiveRamp, 2015; Rubinstein, 2014; Schuster, 2015).The strategies, technologies, and tools of digital political marketing are more complex and far-reaching than anything we have seen before, with further innovations already underway (WARC, 2017). But because most commercial and political digital operations take place below the radar, they are not fully understood by the public. 1

In the following pages, we briefly describe the growth and maturity of digital marketing, highlighting its basic features, key players, and major practices. We then document how data-driven digital marketing has moved into the centre of American political operations, along with a growing infrastructure of specialised firms, services, technologies and software systems. We identify the prevailing digital strategies, tactics, and techniques of today’s political operations, explaining how they were employed during the most recent US election cycle. Finally, we explore the implications of their use for democratic discourse and governance, discussing several recent policy developments aimed at increasing transparency and accountability in digital politics.

Our research for this paper draws from our extensive experience tracking the growth of digital marketing over the past two decades in the United States and abroad, monitoring and analysing key technological developments, major trends, practices and players, and assessing the impact of these systems in areas such as health, financial services, retail, and youth (Chester, 2007; Montgomery, 2007; Montgomery, Chester, & Kopp, 2017). During the 2016 US presidential election, we monitored commercial digital advertising and data use by candidates, parties and special interest groups across the political spectrum. We collected examples of these ads, along with technical and market impact information from the developers of the applications. We also reviewed trade journals, research reports, and other industry documents, and attended conferences that were focused on digital technologies and politics. In the process, we identified all of the major providers of political digital data targeting applications (e.g., Google, Facebook, data clouds, ad agencies) and analysed all their key materials and case studies related to their 2016 operations. The source for much of this work was our ongoing gathering and analysis of cross-sectional commercial digital marketing practices worldwide.

Marriage of politics and commerce

Since the mid-20^th century, advertising has been an increasingly powerful and pervasive presence in US political campaigns, as a growing cadre of ad agencies, public relations firms, and consultants perfected the use of opinion polls, focus groups, and psychographics to reach and influence voters through radio, television, direct mail, and other media outlets (A. Jamieson, 2016; K. H. Jamieson, 1996; Sabato, 1981). With the rise of the internet, campaign operatives began to harness digital technologies and tools to mobilize voter turnout, engage young people, raise money, and support grassroots ground operations (Karpf, 2016; Kreiss, 2016; Tufecki, 2014). Both major political parties in the United States developed large, sophisticated data and digital operations (Kreiss, 2016).

Many of the digital strategies, tools, and techniques employed in the 2016 election were initially developed, deployed, tested, and refined by the commercial sector (Tufecki, 2014).Since its origins in the mid-1990s, digital marketing has operated with a core business model that relies on continuous data collection and monitoring of individual online behaviour patterns (Montgomery, 2011). This system emerged in the United States amid a political culture of minimal government interference, and within a prevailing laissez-faire ethos regarding the internet and new technologies (Barlow, 1996). In the earliest days of the “dot-com boom”, a strong political alliance was forged between the digital media companies and their partners in the advertising and media business, enabling the nascent industry to effectively ward off any attempts to restrain its business operations through privacy regulation or other public policies (Solon & Siddiqui, 2017). As a consequence, the advertising industry played a central role in shaping the operations of platforms and applications in the digital media ecosystem. Digital marketing is now well established and thriving, with expenditures reaching nearly $72.5bn in 2016 for the US alone, and worldwide spending predicted to reach more than $223bn this year (eMarketer, 2017; IAB, n.d.-d).

Ongoing innovations over the years have increased the capacity of data and digital marketing applications. Data collection, analysis, and targeting were further woven into the daily lives of consumers with the rise of social media platforms and mobile devices. Because of the unique role that they play in users’ lives, these platforms are able to sweep up enormous amounts of information, including not only what users post about themselves, but also what is collected from them throughout their daily activities (Smith, 2014). A growing arsenal of software and analytic tools has enhanced the ability of digital media companies and their advertisers to glean valuable insights from the oceans of data they generate (Smith, 2014). Predictive analytics introduced an expanded set of tools for scoring, rating, and categorising individuals, based on an increasingly granular set of behavioural, demographic, and psychographic data (“What is Predictive Intelligence”, 2017). US digital marketers have helped popularise and spur the successful adoption of digital advertising platforms and applications in nearly every geographical location with an internet connection or a link to a mobile device (IAB, n.d.-c). Google, Facebook, and other major players in the digital marketing industry have also developed a global research infrastructure to allow them, and especially their major advertising clients, to make continuous improvements in reaching and influencing the public, and to measure with increasing accuracy the success of their efforts (Facebook IQ, n.d.-a). These developments have created what some observers have called the “surveillance economy” (Singer, 2012).

The growth of data-driven political marketing

Though political campaigns have employed micro-targeting techniques—which use an array of personalised and other data sets and marketing applications to influence the actions of individuals—during the last several election cycles, recent technological innovations and industry advances have created a much more robust system than what was in place in 2012 (IAB, n.d.-b; Rubinstein, 2014). For years, political campaigns have been able to combine public voter files with commercial information from data brokers, to develop detailed and comprehensive dossiers on American voters (Rubinstein, 2014). With recent advances in the advertising technology and data industries, they can now take advantage of a growing infrastructure of specialty firms offering more extensive resources for data mining and targeting voters. Among the new entities are data marketing clouds. Developed by well-known companies such as Adobe, Oracle, Salesforce, Nielsen, and IBM, these clouds sell political data along with an exhaustive amount of detailed consumer information for each potential target, including, for example, credit card use, personal interests, consumption patterns, and TV viewing patterns (Salesforce DMP, 2017).

Some of these massive cloud services also operate what has become a new and essential component for contemporary digital targeting—the data management platform (DMP) (Chavez, 2017). DMPs provide marketers with “centralized control of all of their audience and campaign data” (BlueKai, 2011). They do this by collecting and analysing data about individuals from a wide variety of online and offline sources, including first-party data from a customer’s own record, such as the use of a supermarket loyalty card, or their activities captured on a website, mobile phone, or wearable device; second-party data, information collected about a person by another company, such as an online publisher, and sold to others; and third-party data drawn from thousands of sources, comprising demographic, financial, and other data-broker information, including race, ethnicity, and presence of children (O’Hara, 2016). All of this information can be matched to create highly granular “target audience segments” and to identify and “activate” individuals “across third party ad networks and exchanges”. DMPs are quickly becoming a critical tool for political campaigns (Bennett, 2016; Kaye, 2016, July; Regan, J., 2016).

Facebook and Google now play a central role in political operations, offering a full spectrum of commercial digital marketing tools and techniques, along with specialied ad “products” designed for political use (Bond, 2017). Not surprisingly, these companies have also made generating revenues from political campaigns an important “vertical” category within their ad business (Facebook, n.d.-d; Facebook IQ, n.d.-b; Stanford, 2016). Facebook’s role in the 2016 election was particularly important. With users required to give their real names when they sign up as members, Facebook has created a powerful “identity-based” targeting paradigm, enabling political campaigns to access its more than 162 million US users and to target them individually by age, gender, congressional district, and interests (Facebook, n.d.-b).Its online guide for political campaign marketing urges political campaigns to use all the social media platform tools it makes available to advertisers—including through Instagram and other properties—in order to track individuals, capture their data through various “lead-generation” tactics, and target them by uploading voter files and other data (Facebook, n.d.-a-c-f). The company also employs teams of internal staff aligned with each of the major political parties to provide technical assistance and other services to candidates and their campaigns (Chester, 2017; Kreiss & Mcgregor, 2017). Google heavily promoted the use of YouTube, as well as its other digital marketing assets, during the 2016 US election, reaching out to both major political parties (YouTube, 2017).

The growth and increasing sophistication of the digital marketplace has enhanced the capacities of political campaigns to identify, reach, and interact with individual voters. Below we identify seven key techniques that are emblematic of this new digital political marketing system, providing brief illustrations of how they were employed during the 2016 election.

Cross-device targeting

Getting a complete picture of a person’s persistent “identity” through an “identity-graph” has become a key strategy for successfully reaching consumers across their “omnichannel” experience (use of mobile, TV, streaming devices, etc.) (Winterberry Group, 2016). “Cross-device recognition” allows marketers to determine if the same person who is on a social network is also using a personal computer and later watching video on a mobile phone. Through data “onboarding,” a customer record that may contain a physical and email address is linked through various matching processes, associating it with what is believed to be that individual’s online identification—cookies, IP addresses, and other persistent identifiers (Levine, 2016). Cross-device targeting is now a standard procedure for political initiatives and other campaigns. Voter files are uploaded into the onboarding process, enabling the campaigns to find their targets on mobile devices and at specific times when they may be more receptive to a message (Kaye, 2016, April; L2, n.d.-b). Such granularity of information also enables a more tailored advertisement—so-called “dynamic creative”—which can be changed over time to “deliver very specific messaging” to individuals (Schuster, 2015). Leading cross-device marketing company Drawbridge offered a suite of election services in 2016 that provided campaigns a number of ways to impact voters, including through “Voter-Centric Cross Device Storytelling”, “Political Influencer Identification”, and via “Real-Time Voter Attribution Measurement” (Drawbridge, n.d.).

Programmatic advertising

Programmatic advertising refers to new automated forms of ad buying and placement on digital media using computer programmes and algorithmic processes to find and target a customer wherever she goes. The process can also involve real-time “auctions” that occur in milliseconds in order to “show an ad to a specific customer, in a specific context” (Allen, 2016). The use of programmatic advertising was one of the major changes in political campaign digital operations between 2012 and 2016—“the first time in American History”, according to one ad company, “that such precise targeting has ever been made available at such great scale” (Briscoe, 2017; Kaye, 2015). Programmatic advertising has itself grown in its capabilities to reach individuals, taking advantage of new sources of data to reach them on all of their devices (Regan, T., 2016). In 2016, for example, global ad giant WPP’s Xaxis system—“the world’s largest programmatic and technology platform”—launched “Xaxis Politics”. Capable of “reaching US voters across all digital channels”. the system is said to “segment audiences by hundreds of hot button issues as well as by party affiliation”, including via “real-time campaigns tied to specific real-world events” (Xaxis, 2015). Candidates were able to use the services of a growing list of companies, including Google, Rubicon, AOL, PubMatic, Appnexus and Criteo, that offered programmatic advertising platforms (“Political Campaigns”, 2016; Yatrakis, 2016).

Lookalike modelling

The use of big data analytics enables marketers to acquire information about an individual without directly observing behaviour or obtaining consent. They do this by “cloning” their “most valuable customers”in order to identify and target other prospective individuals for marketing purposes (LiveRamp, 2015). For example, Stirista (n.d.), a digital marketing firm that also serves the political world, offers lookalike modelling to identify people who are potential supporters and voters. The company claims it has matched 155 million voters to their “email addresses, online cookies, and social handles”, as well as “culture, religion, interests, political positions and hundreds of other data points to create rich, detailed voter profiles”. Facebook offers a range of lookalike modelling tools through its “Lookalike Audiences” ad platform. For example, Brad Parscale, the Trump campaign’s digital director, used the Lookalike Audiences ad tool to “expand” the number of people the campaign could target (Green & Issenberg, 2016). Facebook’s “Custom Audiences” product, similarly, enables marketers to upload their own data files so they can be matched and then targeted to Facebook users (Facebook, n.d.-e).

Geolocation targeting

Mobile devices continually send signals that enable advertisers (and others) to take advantage of an individual’s location—through the phone’s GPS (global positioning system), Wi-Fi, and Bluetooth communications. All of this can be done with increasing speed and efficiency. Through a host of new location-targeting technologies, consumers can now be identified and targeted wherever they go, while driving a car, pulling into a mall, or shopping in a store (Son, Kim, & Shmatikov, 2016). A complex and growing infrastructure of geolocation-based data-marketing services has emerged, with specialised mobile data firms, machine-learning technologies, measurement companies, and new technical standards to facilitate on-the-go targeting (Warrington, 2015). The use of mobile geo-targeting techniques played a central role in the 2016 election cycle, with a growing number of specialists offering their services to campaign operatives. For example, L2 (n.d.-a) made its voter file, along with HaystaqDNA modelling data, available for mobile device targeting, offering granular profile data on voters based on their interest in such contested topics as gun laws, gay marriage, voter fraud, and school choice, among others. Conde Nast’s Advance Publications’ election advertising worked with campaigns to append geo-location, profile data, and buying behaviour “to sculpt a very specific voter profile and target down to few hundred readers in a given geo location” (Ellwanger, 2016).

Online video advertising

Digital video, via mobile and other devices, is perceived as a highly effective way of delivering emotional content on behalf of brands and marketing campaigns (IAB, n.d.-a). There are a variety of online video ad formats that provide both short- and long-form content, and that work well for political and other marketing efforts. Progressive political campaign company Revolution Messaging, which worked for the Sanders campaign, developed what it calls “smart cookies” that it says take video and other ad placement “to the next level, delivering precision and accuracy” (Revolution Messaging, n.d.). Google’s YouTube has become a key platform for political ads, with the company claiming that today, voters make their political decisions not in “living rooms” in front of a television but in what it calls “micromoments” as people watch mobile video (DoubleClick, n.d.). According to the company’s political ad services research, mobile devices were used in nearly 60 percent of election-related searches during 2016. Content producers (which it calls “Creators”) on YouTube were able to seize on these election micro-moments to influence the political opinions of potential voters 18-49 (“Letter from the Guest Editors,” 2016).

Targeted TV advertising

Television advertising, which remains a linchpin of political campaign strategy, is undergoing a major transformation, as digital technologies and “addressable” set-top boxes have changed cable and broadcast TV into powerful micro-targeting machines capable of delivering the same kinds of granular, personalised advertising messages to individual voters that have become the hallmark of online marketing. Political campaigns are in the forefront of using set-top box “second-to-second viewing data”, amplified with other data sources, such as “demographic and cross-platform data from a multitude of sources” via information brokers, to deliver more precise ads (Fourthwall Media, n.d.; Leahey, 2016; NCC Media, n.d.). NCC Media, the US cable TV ad platform owned by Comcast, Cox, and Spectrum, provided campaigns the ability to target potential voters via the integration of its set-top box viewing information with voter and other data from Experian and others (Miller, 2017). Deals between TV data viewing companies and organisations representing both Republican- and Democratic-leaning groups brought the “targeting capabilities of online advertising to TV ad buys…bringing what was once accessible only to large state-wide or national campaigns to smaller, down-ballot candidates”, explained Advertising Age (Delgado, 2016).

Psychographic, neuromarketing, and emotion-based targeting

Psychographics, mood measurement, and emotional testing have been used by advertisers for many decades, and have also been a core strategy in political campaign advertising (Key, 1974; Packard, 2007; Schiller, 1975). The digital advertising industry has developed these tools even further, taking advantage of advances in neuroscience, cognitive computing, data analytics, behavioural tracking, and other recent developments (Crupi, 2015). Granular-based messages that trigger a range of emotional and subconscious responses, to better “engage” with individuals and deepen relationships with commercial brands, have become part of the DNA of digital advertising (McEleny, 2016). Facebook (2015), Nielsen, and most leading brands use “neuromarketing” services worldwide, which utilise neuroscience tools to determine the emotional impact of advertising messages. There is a growing field, recently promoted by Google, of “Emotion Analytics” that takes advantage of “new types of data and new tracking methods” to help advertisers “understand the impact of campaigns—and their individual assets—on an emotional level…” (Kelshaw, 2017). Scholars have identified that the use of “psychological targeting” in advertising enables the influencing of large groups of people by “tailoring persuasive appeals to the psychological needs” of the specific audiences (Matz, et al, 2017). Data company Experian Marketing Services for political campaigns offered data that weaved together “demographic, psychographic and attitudinal attributes” to target voters digitally. Experian claims its data enables campaigns to examine a target’s “heart and mind” via attributes related to their “political persona” as well as “attitudes, expectations, behaviours, lifestyles, purchase habits and media preferences (Experian, 2011, 2015). One of the most well publicised and controversial players in the 2016 election was Cambridge Analytica (CA), a prominent data analytics and behavioural communications firm that claimed to be a key component in Donald Trump’s victorious campaign. The company used a “five-factor personality model” aimed at determining “the personality of every single adult in the United States of America” (Albright, 2016; Kranish, 2016).Known as OCEAN, the model rated individuals based on five key traits: openness, conscientiousness, extroversion, agreeableness, and neuroticism. Drawing from digital data, voter history, and marketing resources supplied by leading companies, including Acxiom, Experian, Nielsen, GOP firm Data Trust, Aristotle, L2, Infogroup, and Facebook, CA was able to develop an “internal database with thousands of data points per person”. Its research also identified key segments that were considered “persuadable”, and shaped the advertising content placed “across multiple digital channels (with the most effective ads also appearing on television) (Advertising Research Foundation, 2017; Nix, 2016). The strategy was based on developing messages that were tailored to the vulnerabilities of individual voters (Nix, 2016; Schwartz, 2017). CA has become the subject of much scrutiny and debate, and itself has made conflicting claims, with critics raising concerns over its techniques and expressing scepticism about the extent of its impact (Confessore & Hakim, 2017; Karpf, 2017). However, the company’s work was sufficiently convincing to the leading advertising industry research organisation, the Advertising Research Foundation (2017, March), that it honoured the firm with a “Gold” award in 2017 under its “Big Data” category.

Discussion

The above description provides only a brief overview of the data-driven marketing system that is already widely in use by candidate and issue campaigns in the United States. The increasingly central role of commercial digital marketing in contemporary political campaigns is reshaping modern-day politics in fundamental ways, altering relationships among candidates, parties, voters, and the media. We acknowledge that digital technologies have made important positive contributions to the vibrancy of the political sphere, including greatly expanding sources of news and information, significantly increasing opportunities for citizen participation, and empowering people from diverse backgrounds to form coalitions and influence policy. The same tools developed for digital marketing have also helped political campaigns substantially improve voter engagement, enhance their capacities for “small-donor” fundraising, and more efficiently generate turnout (Moonshadow Mobile, n.d.; Owen, 2017). However, many of the techniques we address in this paper raise serious concerns—over privacy, discrimination, manipulation, and lack of transparency.

Several recent controversies over the 2016 election have triggered greater public scrutiny over some of the practices that have become standard operating procedure in the digital media and marketing ecosystem. For example, “fake news” has a direct relationship to programmatic advertising, the automated system of “intelligent” buying and selling of individuals and groups (Weissbrot, 2016). These impersonal algorithmic machines are focused primarily on finding and targeting individual consumers wherever they are, often with little regard for the content where the ads may appear (Maheshwari & Isaac, 2016). As a consequence, in the middle of the 2016 election, many companies found themselves with ads placed on “sites featuring pornography, pirated content, fake news, videos supporting terrorists, or outlets whose traffic is artificially generated by computer programs”, noted the Wall Street Journal (Nicas, 2016; Vranica, 2017). As a major US publisher explained in the trade publication Advertising Age,

Programmatic’s golden promise was allowing advertisers to efficiently buy targeted, quality, ad placements at the best price, and publishers to sell available space to the highest bidders…. What was supposed to be a tech-driven quality guarantee became, in some instances, a “race to the bottom” to make as much money as possible across a complex daisy chain of partners. With billions of impressions bought and sold every month, it is impossible to keep track of where ads appear, so “fake news” sites proliferated. Shady publishers can put up new sites every day, so even if an exchange or bidding platform identifies one site as suspect, another can spring up (Clark, 2017).

Criticism from news organisations and civil society groups, along with a major backlash by leading global advertisers, led to several initiatives to place safeguards on these practices (McDermott, 2017; Minsker, 2017). For example, in an effort to ensure “brand safety”, leading global advertisers and trade associations demanded changes in how Google, Facebook and others conduct their data and advertising technology operations. As a consequence, new measures have been introduced to enable companies to more closely monitor and control where their ads are placed (Association of National Advertisers, 2017; Benes, 2017; IPA, 2017; Johnson, 2017; Liyakasa, 2017; Marshall, 2017; Timmers, 2015).

The Trump campaign relied heavily on Facebook’s digital marketing system to identify specific voters who were not supporters of Trump in the first place, and to target them with psychographic messaging designed to discourage them from voting (Green & Issenberg, 2016). Campaign operatives openly referred to such efforts as “voter suppression” aimed at three targeted groups: “idealistic white liberals, young women and African Americans”. The operations used standard Facebook advertising tools, including “custom audiences” and so-called “dark posts”—“nonpublic paid posts shown only to the Facebook users that Trump chose” with personalised negative messages (Green & Issenberg, 2016). Such tactics also took advantage of commonplace digital practices that target individual consumers based on factors such as race, ethnicity, and socio-economic status (Google, 2017; Martinez, 2016; Nielsen, 2016). Civil rights groups have had some success in getting companies to change their practices. However, for the most part, the digital marketing industry has not been held sufficiently accountable for its use of race and ethnicity in data-marketing products, and there is a need for much broader, industry-wide policies.

Conclusion

Contemporary digital marketing practices have raised serious issues about consumer privacy over the years (Schwartz & Solove, 2011; Solove & Hartzog, 2014). When applied to the political arena, where political information about individuals is only one of thousands of highly sensitive data points collected and analysed by the modern machinery of data analytics and targeting, the risks are even greater. Yet, in the United States, very little has been done in terms of public policy to provide any significant protections. In contrast to the European Union, where privacy is encoded in law as a fundamental right, privacy regulation in the US is much weaker (Bennett, 1997; Solove & Hartzog, 2014; U.S. Senate Committee on Commerce, Science, and Transportation, 2013). The US is one of the only developed countries without a general privacy law. As a consequence, except in specific areas, such as children’s privacy, consumers in the US enjoy no significant data protection in the commercial marketplace. In the political arena, there is even less protection for US citizens. As legal scholar Ira S. Rubinstein (2014) explains, “the collection, use and transfer of voter data face almost no regulation”. The First Amendment plays a crucial role in this regard, allowing the use of political data as a protected form of speech (Persily, 2016).

The political fallout over the how Russian operatives used Facebook, Twitter, and other sites in the 2016 presidential campaign has triggered unprecedented focus on the data and marketing operations of these and other powerful digital media companies. Lawmakers, civil society, and many in the press are calling for new laws and regulations to ensure transparency and accountability for online political ads (“McCain, Klobuchar & Warner Introduce Legislation”, 2017). The U.S. Federal Election Commission, which regulates political advertising, has asked for public comments on whether it should develop new disclosure rules for online ads (Glaser, 2017). In an effort to head-off regulation, both Facebook and Twitter have announced their own internal policy initiatives designed to provide the public with more information, including what organisations or individuals paid for political ads and who the intended targets were. These companies have also promised to establish archives for political advertising, which would be accessible to the public (Falck, 2017; Goldman, 2017; Koltun, 2017). The US online advertising industry trade association is urging Congress not to legislate in this area, but to allow the industry to develop new self-regulatory regimes in order to police itself (IAB, 2017). However, relying on self-regulation is not likely to address the problems raised by these practices and may, in fact, compound them. Industry self-regulatory guidelines are typically written in ways that do not challenge many of the prevailing (and problematic) business practices employed by their own members. Nor do they provide meaningful or effective accountability mechanisms (Center for Digital Democracy, 2013; Gellman & Dixon, 2011; Hoofnagle, 2005). It remains to be seen what the outcome of the current policy debate over digital politics will be, and whether any meaningful safeguards emerge from it.

While any regulation of political speech must meet the legal challenges posed by the First Amendment, limiting how the mining of commercial data can be used in the first place can serve as a critically important new electoral safeguard. Advocacy groups should call for consumer privacy legislation in the US that would place limits on what data can be gathered by the commercial online advertising industry, and how that information can be used. Americans currently have no way to decide for themselves (such as via an opt-in) whether data collected on their finances, health, geo-location, as well as race or ethnicity can be used for digital ad profiling. Certain online advertising practices, such as the use of psychographics and lookalike modelling, also call for rules to ensure they are used fairly.

Without effective interventions, the campaign strategies and practices we have documented in this paper will become increasingly sophisticated in coming elections, most likely with little oversight, transparency, or public accountability. The digital media and marketing industry will continue its research and development efforts, with an intense focus on harnessing the capabilities of new technologies, such as artificial intelligence, virtual reality, and cognitive computing, for advertising purposes. Advertising agencies are already applying some of these advances to the political field (Facebook, 2016; Google, n.d.-a; Havas Cognitive, n.d.). Academic scholars and civil society organisations will need to keep a close watch on all these developments, in order to understand fully how these digital practices operate as a system, and how they are influencing the political process. Only through effective public policies and enforceable best practices can we ensure that digital technology enhances democratic institutions, without undermining their fundamental goals.

This paper is part of 'A Manchurian candidate or just a dark horse? Towards the next generation of political micro-targeting research’, a Special issue of the Internet Policy Review.

Introduction

As political campaigns compete, they try to outsmart each other by all sorts of actions: from dropping witty puns during a televised debate, to strategically knocking on doors and convincing voters. Technological innovation can help political parties improve the effectiveness of their campaigns. By using technology to collect, process, and analyse information about voters, campaigns can improve their knowledge about the electorate. Subsequently, technology can extend campaigns' capabilities of targeting specific groups with tailored messages resulting in more efficient campaigning. We call this phenomenon 'political behavioural targeting' (PBT).

Several scholars have researched political behavioural targeting in the US context (e.g., Kreiss, 2012, 2016; Nielsen, 2012; Hersh, 2015). However, the US differs in several obvious ways from most European countries. One can imagine that differences in electoral systems, privacy laws, and party financing influence campaigns’ ability to collect, process, and use personal voter data. Therefore, the findings from these studies do not necessarily apply to European countries. As there is little research in a European context, it remains unclear to what extent and how campaigns in a multiparty democracy, such as the Netherlands, use PBT-techniques. Also, it is unclear if and why there are differences between parties. In line with Colin Bennett (2016, p. 261), we wonder: "can political parties campaign in Europe as they do in North America?"

Such a question is relevant, as some scholars fear that the use of data and targeting techniques hinders public deliberation (Gorton, 2016), weakens the mandate of elected officials (Barocas, 2012), has negative effects on citizens’ privacy (Howard, 2006; Rubinstein, 2014; Tene, 2011), and enables campaigns to send tailored messages directly to citizens, thereby avoiding scrutiny from journalists (Jamieson, 2013). As a result, campaigns can potentially make opposite promises to different people, without anyone noticing.

This article sheds light on how Dutch political campaigns adopt and use PBT-techniques. Through interviews with campaign leaders, using a grounded theory approach, we answer the following overarching research question: What barriers and facilitators for the adoption and use of PBT-techniques do Dutch political parties perceive?

Theoretical framework

We will first summarise innovations in political campaigns over time, leading up to the advent of political behavioural targeting. Then, we identify the factors influencing the adoption of PBT on a campaign team level. Finally, we explore the factors that can shape the adoption of PBT on the level of national systems.

Innovations in political campaigns

Political campaigns have continuously been adapting to technological developments. Pippa Norris (2000) describes how the advent of television and the shift from partisan newspapers to national television news triggered a process of modernisation in the way political campaigns operated. Notable consequences of this shift were the adoption of a media-centred strategy in order to set the agenda, the rise of political marketing, the collection and use of data (such as opinion polls) to "shape, fine-tune and monitor campaign efforts" (Blumler, Kavanagh, and Nossiter, 1996; p. 53). Another shift came with the internet and the new possibilities for party-voter interaction that came along with the medium, which led campaigns to a new stage of the modernisation process: the postmodern campaign (Norris, 2000).

It would be an oversimplification to point to 'the internet' as a game-changer in political communication, because of the rapidly changing nature of the internet itself. As David Karpf (2012, p. 640) notes: "the internet of 2002 has important differences from the internet of 2005, or 2009, or 2012". Accordingly, much more than the advent of the internet itself, it is the advent of social media such as Facebook (2004), YouTube (2005) and Twitter (2006) which provided political campaigns with new ways of communication with the electorate (e.g., Gibson & McAllister, 2011; Conway, Kenski, & Wang, 2015; Vaccari, 2012). Together with companies such as Google, whose core business is actually not its well-known search engine but rather its advertisement business, social media not only facilitate new ways of communication, but also the tracking and collection of behavioural data of internet users (Zuiderveen Borgesius, 2016). This technique ("behavioural targeting") originates from the advertisement business. Ad agencies monitor people's online behaviour and combine this information with consumer data provided by data brokers, to target them individually with tailored ads (Turow, 2011, p.75). When applying this concept to the political realm, we can dub this phenomenon as political behavioural targeting (PBT).

Of course, PBT is not about selling products but about winning votes. And political campaigns have different means to do so than advertisement agencies have (e.g. canvassing efforts); which means that PBT happens offline as well as online. We distinguish traditional canvassing from PBT-canvassing if campaigns are able to process information about individual conversations (such as the voter's likelihood to vote for a party or her most important voting consideration), and subsequently use that information to gain strategic insights about the electorate and/or to target the voter at a later stage with a tailored message, while skipping the 'wrong' doors in a neighbourhood (Kreiss, 2016; Nielsen, 2012).

Arguably, the use of PBT can be seen as the latest step within the modernisation of political campaigns. However, as we have seen in earlier phases, not all parties in all countries adopt new techniques at the same pace and rate. Below, we identify the factors influencing the adoption of PBT. We organise these factors at two levels: (1) the individual campaign around a candidate/party and (2) the national system (i.e., the electoral system, regulatory framework, and culture). This translates into the model shown in Figure 1, which will be elaborated on in the next paragraphs.

Figure 1: Factors influencing the adoption of PBT

The campaign team level

In his extensive research of US political campaigns, Daniel Kreiss (2016) identified four factors concerned with technological innovation within political campaigns. There are resource factors, such as campaign budgets and the number of volunteers a campaign can employ; infrastructural factors, such as technological tools or skills within the organization; organizational factors, such as organisational culture and structure; structural electoral cycle factors, such as election results. Building upon Kreiss' factors, we add an additional four (one campaign team level factor and three system level factors) to examine the use of PBT. On a campaign team level, the factor is ethical and legal concerns, such as normative reservations towards PBT. On a system level, the factors are electoral context, regulatory framework,and culture (discussed below). These new factors were identified through a review of literature about innovation in data-driven political campaigning techniques (e.g., Anstead, 2017; Kreiss, 2016; Jungherr, 2016; Hersh, 2015; Nielsen, 2012), and literature about (hybridisation of) campaign evolvement (e.g., Lijphart, 2012; Plasser & Plasser, 2002; Karlsen, 2010; Norris, 2000).

Resource factors

The main elements within this factor that could influence the extent to which campaigns can use PBT-techniques are: the budget and the effort needed to carry out a PBT-operation. A large budget enables campaigns to hire skilled personnel, acquire data, or buy targeted ads. The same dynamic applies to the number of volunteers a campaign can mobilise: having a lot of them facilitates a campaign in collecting data by canvassing, and sending potential voters targeted messages (the use of volunteers, of course, is dependent on their skills). Having a small budget and few volunteers, consequently, can be a barrier for campaigns because it bars them from acquiring the same amount of capabilities or from carrying out an operation on a large scale. This is in line with normalization theory (Margolis & Resnick, 2000), according to which the possibilities of the internet will not upset traditional power structures, but will rather develop along traditional lines as in the 'offline world'.

We can also view PBT as a means of using a campaign's resources as efficient as possible, to ensure parties do not spend money and effort on voters who will vote for another party anyway, or on citizens who will not vote altogether. Then, parties with limited resources could be more inclined to use PBT to not waste precious money, time, and labour. This is in accordance with the idea of equalization, which views the internet as an empowering tool for smaller parties due to its low costs and its new ways of direct communication with the electorate (Margolis, Resnick, & Levy, 2003; Bimber & Davis, 2003; Stanyer, 2010). A meta-analysis found evidence for the existence of both normalisation and equalisation in election campaigns (Strandberg, 2008). The occurrence of either process can differ per country and is dependent on several contextual factors, which will be discussed later on.

Organisational factors

The elements in this factor are about how campaign leaders perceive campaigning. Do they rely on proven best practices from previous campaigns or is there a culture of innovation? John Padgett and Walter Powell (2012) describe the concept of network folding. Applied to the political realm, this entails the extent to which campaigns employ skilled personnel from non-political sectors and to integrate that expertise into their existing institutions. An example is the hiring of Google engineer Stephanie Hannon as chief technology officer by the Clinton campaign (Easton, 2015). The 'cognitive diversity' following from network folding can lead to creative ideas (De Vaan, Stark, & Vedres, 2015). Furthermore, the organisational structure can be expected to resemble the way the campaign perceives PBT. A campaign with an autonomous data department is probably more prone to rolling out a PBT-strategy than a campaign that sees 'data' as only one of the many tasks of a communication staffer. Also, a change in leadership can be a facilitator for innovation (Gibson & Römmele, 2001).

Infrastructural factors

Elements are the technological tools available to campaigns, which enable them to roll out a PBT-operation. For instance, such tools might assist volunteers in the field by enabling them to collect data. They can be developed in-house or outsourced; in fact, there are specialised third party consultancies, that offer off-the-shelf tools, which in turn allow campaigns to employ innovative technology even though the campaigners do not have any technical expertise.

Structural electoral factors1

The actions of rival campaigns fall under the umbrella of structural electoral factors. A successful PBT-campaign of a rival can facilitate innovation in other campaigns, especially if those other campaigns themselves look back at an unsuccessful election. This connects with the 'critical event' (Kreiss, 2016), such as losing an election that should have been won, or with the experience of an 'external shock', which can be an incentive for professionalisation (Gibson & Römmele, 2001).

A second element influencing campaigns' likelihood to use PBT-techniques, is issue ownership (Petrocik, 1996), and the subsequent statements of party candidates propagating standpoints of the party. A political campaign 'caught' using privacy-infringing PBT-techniques, while its candidates present themselves as privacy champions, is likely to come across as hypocritical. Being perceived as such should be avoided, considering the negative electoral consequences following political-ideological hypocrisy perpetrated by politicians (Bhatti, Hansen, & Olsen, 2013).

Ethical and legal concerns

Elements within this factor consist of ethical and legal restrictions on how campaigns operate. For example, a political party could believe that PBT is ethically wrong as it infringes on citizens' right to privacy, and citizens' autonomy to form their own opinions. As a result, the party 'self-regulates' and refrains from using campaigning techniques violating its ethical beliefs.

Another element is the legal uncertainty that occurs when a campaign does not know how to behave in accordance with data protection and election laws, because of a lack of internal expertise. Such confusion can result in differences in the actions taken by comparable actors (e.g. Raskolnikov, 2017). Legal uncertainty can lead to 'overcompliance', which can be seen as a barrier towards the adoption of PBT-techniques, or to 'undercompliance', which facilitates the adoption of PBT-techniques (Calfee & Craswell, 1984). For instance, Anstead (2017) notes how parties felt disadvantaged by targeting possibilities facilitated by the perceived undercompliance with UK campaign finance law during the 2015 general elections.

The system level

Aside from campaign level factors, we look at contextual factors as well. These factors may limit the extent to which (US-American) campaigning techniques can be adopted in other countries (Karlsen, 2010). Therefore, we add three new contextual factors to our model. We expect that the electoral system, the regulatory system, and the culture of a democracy influence the extent to which the campaign team level factors are applicable. Below, we explore how the adoption of PBT-techniques can be influenced by properties of different systems. We will later apply our model (see Figure 1) to one specific case.

Electoral system

The three dominant electoral systems are first-past-the-post (FPTP), proportional representation (PR), and two-round (TR) (Birch, 2001, 2003). How these systems function, can influence how campaigns are run. The FPTP-system, first, can lead to an overvaluation of some key districts. Such districts sometimes 'swing' to one party and sometimes to another party, whereas other districts go to the same party in each election. As an effect, campaigns in a FPTP-system are inclined to spend a disproportionate amount of money and labour in these key districts in the hope of swinging the election their way (e.g., Anstead, 2017; Lipsitz, 2004). The PR-system, second, does not favour a select group of voters in a few key districts (Plasser & Plasser, 2002). This is especially true when the PR-system consists of only one district, in which every vote counts equally. As a result, campaigns have to spread their means more equally over the country. The TR-system, third, makes for a relatively unpredictable campaign, since it often is unclear which candidates will make it to the second round. Furthermore, the TR-system makes it important for campaigns to collect the votes of the supporters of the losing candidates of the first round. Therefore, campaigns should not only focus on their own base but other candidates' bases as well (Blais & Indridason, 2003). This has consequences for PBT, since campaigns should not only correctly classify potential voters as their own, but the other voters as well in order to target them in the next round.

A different aspect of electoral systems that influences how a campaign is run, is the degree of fractionalisation in a democracy (Duverger, 1959; Lijphart, 2012; Wang, 2012). FPTP-systems favour relatively few candidates/parties. PR-systems, in contrast, enable a large number of parties to run in an election. The first round of a TR-system can consist of many different candidates. As a result, campaigns that operate in a PR or a TR-system are less likely to launch attack campaigns against competitors. This is because PR-systems generally require a coalition of parties working together after the elections (Plasser & Plasser, 2002). And in a TR-system, campaigns should not mistreat competing candidates too much because winning campaigns have to court the bases of losing candidates in order to win the second round. Furthermore, in a highly fractionalised democracy, parties represent different (minority) groups within the electorate. This results in a high risk of 'mistargeting', in which campaigns approach a member of group A with appeals for a member of group B. Mistargeting can lead to voters penalising the campaign for their mistake (Hersh, 2013). These contextual circumstances may call for different PBT-strategies.

Regulatory framework

We distinguish strictly regulated campaign environments, moderately regulated campaign environments and minimally regulated campaign environments (Plasser & Plasser, 2002). Strictly regulated environments are characterised by "severe restrictions on the contact and communication between candidates and their constituencies" (p. 137). Moderately regulated environments typically focus on regulating access to TV-advertising and campaign funding. Minimally regulated environments impose few regulatory restrictions on political campaigns. It may be infeasible to implement innovative PBT-techniques in strictly regulated environments. Legal uncertainty can play a role on a system level too (e.g. because of a gap in the law).

Culture

Differences in the adoption and use of innovative PBT-practices can also be influenced by the culture or tradition in a democracy. For example, turnout culture is important because campaigns operating in countries where turnout is high will focus more on convincing voters on getting out the vote than campaigns operating in a low-turnout culture. In a low-turnout culture, campaigns sometimes target specific groups of voters (e.g. the elderly, who are more likely to turn out) more than other groups (e.g. the poor), who are unlikely to turn out (Herrnson, 2001). The turnout culture can influence the data campaigns collect on someone (and how campaigns tailor their messages), because a campaign message meant to convince someone typically leans on more data than a message meant to mobilise a voter does. Furthermore, cultural norms can dictate the strategy of political campaigns. In Japan, for example, posting dark post attack ads, such as the 'super predator' ad Trump launched against Clinton (Green & Issenberg, 2016), is improbable because of the cultural convention of averting direct conflict (Plasser & Plasser, 2002).

System level context is likely to affect campaign level factors. A campaign operating in a multiparty PR-system needs to pour more resources into identifying potential supporters than a campaign in a FPTP-system. After all, identifying potential Republicans or Democrats is easier than identifying potential voters in a ten-party race. Moreover, other than in US campaigns, most European campaigns are unable to access voter registration files provided by an electoral register. In many countries citizens can just show up at the voting booth, which means that the whole act of ‘registering’ for voting, as it is the case in the US, does not exist. Since Hersh (2015) has found that voter lists are one of the most valuable pieces of data to US campaigns, this principal unavailability, or non-existence, of such data poses a challenge for the PBT-capabilities of campaigns. But this challenge should by no means imply a PBT-operation in Europe is impossible. We would argue that while the lack of access to voter lists makes it very difficult to achieve the same level of granularity when engaging in PBT as in the US, by using other commercially available or self-collected data, it can be possible to come reasonably close to the desired level (which may be more modest for European campaign leaders than for American ones). The extent to which there is an actual difference between the degree of granularity between US and European campaigns, however, is outside of the scope of this paper, as we focus on perceptions and strategies of campaign leaders.

System level context also affects infrastructure. For instance, should the groundwork be spread equally across the nation, or focused on a number of battleground states? Also, a campaign operating in a heavily regulated context is likely to encounter some legal barriers. For example, because of campaign financing regulations (may influence resource factors), and data protection regulations (infrastructural factors). The absence of regulations, conversely, can facilitate PBT. Cultural context, finally, can influence campaigns' ethical considerations regarding PBT. Campaigns operating in a culture that favours privacy, for example, can be expected to avoid (or use less-invasive) PBT-techniques than campaigns run in a culture in which privacy is less important. In sum, there are several factors, both on campaign and system level, which can form a barrier or facilitate the extent to which campaigns are able to use PBT-techniques and how they use them.

Extending existing research to a European context, we have developed and will apply an improved model (applicable in different electoral contexts) to analyse barriers and facilitators to innovative PBT-practices by political campaigns. As the context of the research case differs from the US, we expect to contribute to the framework and to shed light on how contextual factors influence innovation of political campaigns. Furthermore, in answering our research question, we provide insight into the way political campaigns in a multiparty democracy organise, communicate and innovate. Given these considerations, our key question is: What barriers and facilitators for the adoption and use of PBT-techniques do Dutch political parties perceive?

Method

This study focuses on campaigns in the Netherlands because of the national elections taking place in the research period (15 March 2017), the advanced technological infrastructure (Coy, 2015), and the interesting contextual factors. The Dutch electoral system is one of open list proportional representation (PR), in which all members of parliament come from one nationwide district (Lijphart, 2012). This means that in the Netherlands, every vote counts equally. Moreover, the system of PR (and the very low de-facto threshold) enables a relatively large number of political parties to run in an election. 28 parties participated in the 2017 national election ("Partijen nemen deel", 2017). Of these parties, 13 actually gained a seat in parliament ("Officiële uitslag", 2017).

The Dutch national elections have a relatively high turnout: around 80% in the previous two elections ("Officiële uitslag", 2017). But where the US presidential campaigns can spend hundreds of millions of dollars (Narayanswamy, Cameron, & Gold, 2017), the Dutch campaign with the biggest funds (VVD) has no more than 5 million dollars to spend. And even if the budgets were sufficiently large, the question is whether voter data would be usable for a political campaign. The Dutch data protection law categorises political preference as sensitive personal data. This means that campaigns are only allowed to process such information if the potential voter explicitly gives permission to do so.

Finally, as party membership in the Netherlands steadily decreases, political campaigns can rely less on their members to do labour-intensive tasks (such as canvassing). In 2016, the number of party members of all political parties combined, was at its lowest point since the second world war. Although this number has picked up slightly since, party membership is still quite low ("Membership Dutch parties still low", 2017).

After approval from the ethical committee of the University of Amsterdam, we carried out eight in-depth interviews with campaign leaders. We interviewed 11 campaign leaders, belonging to eight political parties in total (three interviews were double-interviews). In addition to this, we held two background interviews (with one local campaigner for the municipality of Amsterdam and one political consultant offering PBT-services). The eight elite-interviews on average lasted 53 minutes. Two were conducted by phone, the others face-to-face. We took a qualitative research approach for several reasons: the small group of people concerned with the coordination of political campaigns in the Netherlands, the lack of knowledge on this topic in the Netherlands, and because interviewing is a suitable method for understanding the mechanisms behind and perceptions of a phenomenon (Boeije, 2005). As we want to understand how campaigns see PBT, what they are actually doing, and how they perceive possible barriers and facilitators to the adoption and use of PBT-techniques, the interview is a suitable data-collection method. Using an interview guide (see appendix A), we held semi-structured interviews, allowing follow-up questions.

Interviewees

We selected the interviewees via purposive sampling. Campaign leaders qualified for an interview when they had a coordinating role in the campaign and were campaigning for a party that gained at least one seat in the 2012 national parliamentary elections. Eleven campaigns satisfied this second criterion (see Table 1). We contacted interviewees via email, explaining the objective of the study. The interviewees signed an informed consent document before the interview started. We also promised the campaign leaders anonymity, and confidentiality until after election day (15 March 2017). By doing so, we tried to provide the interviewees with a safe environment in which they felt like speaking freely, and without concern of somehow 'leaking' strategic information. Because no information would become public before election day, the risk that interviewees might provide biased information due to a strategic agenda was minimised. Another large advantage of interviewing the campaign leaders before the elections took place, is the prevention of hindsight bias by the interviewees. Unfortunately, we were unable to convince three parties to comply (VVD [right on the political spectrum], PVV [right-wing nationalist party], PvdD [Party for the animals; left-wing]). These three parties were unwilling to cooperate, either because they still found the risk of leaking their strategy too large, or they did not offer an explanation.

Analysis

Using a grounded theory approach, this study has passed four phases: the exploration phase, the specification phase, the reduction phase, and the integration phase (Wester, 1995). In the exploration phase, two background interviews took place (with a campaigner for the municipality of Amsterdam and with a political consultant offering PBT-services). These were coded using AtlasTI, 'tentatively labelling' relevant information (Glaser, 1978). Thereafter, the first interviews with campaign leaders took place. These were transcribed and open-coded. Furthermore, fellow researchers also coded these interviews and discussed the content (peer debriefing). In the next phase, new interviews took place and the data was subject to axial coding. The first dimensions were identified (e.g., what forms a barrier and what facilitates the use of PBT-techniques?). The reduction phase saw the emergence of the core category (innovation). In the integration phase, we completed the conceptual framework, finalised our analysis, and had the campaign leaders approve the quotes used (member checking). This means the campaign leaders agreed with the way they were quoted, and with the publishing of the names of the political parties. Member checking increased the willingness of campaign leaders to cooperate with the study.

Results

We first describe the field: to what extent do campaigns use PBT-techniques? Then we explain differences between parties by focusing on the five campaign level factors concerning the use of PBT (resource, infrastructural, organisational, structural electoral cycle factors, and ethical and legal concerns). Finally, we zoom out to the system level and discuss the influence of contextual factors (electoral system, regulatory framework and culture) on the adoption of PBT-techniques.

PBT in Dutch campaigns

As campaigns in the Netherlands can have recourse to relatively detailed public census data and detailed election results, all campaigns, to some extent, adopt a PBT-approach. Furthermore, Facebook is an important tool for all parties, but the parties differ in how they use Facebook's capabilities. Some parties occasionally post content targeted to broad age groups, while other campaigns frequently post content tailored to more specific groups. Two campaigns stand out, as they have developed their own PBT-tools, which they can use to continuously refine their knowledge of the electorate. We will now use our model to explain the differences between campaigns.

Resource factors

All campaigns cite financial costs as a barrier. Table 2 shows that budgets are modest, and differ between parties.

Source: parties' annual financial reports, on file with authors.

These small budgets form a barrier for the cooperation with expert political consultants (such as Blue State Digital) to enhance their PBT-operations. Parties refer to the financial costs as the main reason not to hire consultants. Campaign leader 1 of the Social Democratic Party (PvdA) explains why he does not work with Blue State Digital (BSD):

Their system is very expensive, that's a factor. And you need the people to carry out the work for you. In an ideal world, such a cooperation would be really cool though."

Liberal democrat party D66 agrees: "because it costs a lot of money and we don't have that kind of money. And if we spend it on a consultant, we can't spend it on the campaign itself."

The same barrier appears when campaigns speak about other technological means, such as canvassing apps, allowing campaigns to directly process information from canvassers. Christian Democrats CDA, for instance, would like such an app. Campaign leader 8: "yes, but that would demand a financial investment that we can't afford."Green Party GroenLinks has a contrasting perspective: "I believe it usually costs around €100,000 to build an app such as our own. (..) We, however, paid our programmers two crates of beer and 40 pizzas." Several facilitators help GroenLinks and also socialist party SP to overcome this barrier of financial costs. First: the personal network of the campaign leader. This facilitator is especially prominent for GroenLinks, where campaign leader 7 employs his own network to optimise the BSD-systems, but also to help him with setting up other parts of the campaign:

We had to adjust it [the BSD system] somewhat for the Netherlands. The people with whom I did so, Swedish folks ... they are simply a little network of people of around my own age, and some people who are a bit older and have already set up a similar campaign in their own countries. A guy who set up the grassroots organization for Trudeau, for example, he's a couple of years older than I am, but I Skype with him to talk about how I should handle certain things.

Campaign leader 7's personal network plays (or at least played)an important role in cheaply setting up technological tools and creating content:

Through the network, I'm aware of the crowdfunding streams for a normal campaign. (..) I'm meeting a friend tomorrow, who has experience with mail flows. (..) I have a network of volunteering writers, poets, freelance journalists who write for us for free. (..) So partly, I just have a good personal network.

However, as PvdA notes, having lots of data is of no use if you don't have the capacity to use it. GroenLinks tries to overcome this barrier by organising their campaigns, to some extent, in a citizen-initiated manner (Gibson, 2015). A citizen-initiated campaign (CIC), devolves "power over core tasks to the grassroots" (p.183). As campaign leader 7 puts it:

Grassroots is about creating an infrastructure to enable as many sympathizers as possible to volunteer as canvassers on a large scale. So voter contact on a large scale, but also – and that's Bernie's [Sanders] lesson – to have places in which a few people make stuff by themselves without us having any control over it. (..) Embracing people's creativity without managing it.

SP has less need of a citizen initiated campaign, because of their relatively large number of active party members. "The big difference [with GroenLinks] is the fact that we already have the volunteers. Many other parties lack the numbers. We have thousands of party members who gladly canvass for us two weekdays and on Saturday as well." [Campaign leader 11, SP].

Infrastructural factors

Having a good infrastructure allows campaigns to actually collect data and send tailored messages. What kind of PBT-infrastructure can parties rely on and how does it facilitate their use of PBT?

All campaigns use the PBT-infrastructure Facebook offers, although some more than others. Nearly all campaigns use its lookalike audiences function to find new potential voters. Campaign leader 8: "we search for profiles of people who look like the ones who’ve already liked our Facebook page, and then serve them with advertisements." Campaigns also look at people who like pages that are close to the values of the political parties. Christian party ChristenUnie, for instance, tries to target voters who like the page of evangelical broadcaster EO. So does the Calvinist Political Party SGP, which tries to find out people’s interests on Facebook: "For example… farming, or Israel, off the top of my head; you try to approach people along the lines of their interest, or the region in which they reside." [Campaign leader 9]

Some campaigns also employ ‘dark posts’, a Facebook function that enables campaigns to opaquely target specific audiences, while its messages are not visible to untargeted Facebook users. Campaign leader 1 exemplifies:

We’ve managed to get something done related to gas extraction in Groningen. It doesn’t make sense to share that on the national Facebook page, because it was only important news locally. So we put out a dark post, only for Groningen residents. Sometimes we can specify it even more.”

Using Facebook for PBT-purposes, campaigns do not actually gather or own data themselves. There are a few campaigns that do gather their own data, by using canvassing apps. Campaign leader 7:

We use the election results per voting location and use that information to establish the GroenLinks mindedness of a neighbourhood. Then we can prioritise which addresses to visit and which to ignore. When we visit addresses, our volunteers use the app to answer the following questions: 1. Is anyone home? 2. Does she want to talk? 3. Is she going to vote? 4. Is she planning to vote for GroenLinks? 5. What is the most important theme to her? 6. How GroenLinks minded was she? If she considers to vote for GroenLinks, two questions follow: 1. Do you want to stay informed of our campaign by e-mail? 2. Can I have your phone number, so we can ask you to do canvassing talks?”

The GroenLinks app facilitates large scale collection of information about people's political preferences, thereby informing strategic decisions. Also, the personal data can facilitate accurate PBT on an individual level. The secondary objective of the app is to provide an infrastructure for volunteers to campaign on their own terms, whenever they feel like doing so:

Our app, built by hackers, enables others to campaign for us. (..) Someone in [small town] Lutjebroek can install our app and go ahead and work for our campaign. No campaign leader needed. [Campaign leader 7]

Some campaigns monitor the visitors of their own websites. Campaign leader 1: “What are people searching for on our website, how do they get to our website, how much time do they spend, (..) which button should you colour red? How does that work?” At the time of the interview, CDA was not yet tracking their website visitors, but: “we’ve just migrated to a new website, on which we want to start collecting more data on our visitors. I’m curious what kinds of people are visiting the website. And what kinds of people don’t, and therefore have to be reached through different channels.”

SP has built a system which combines previous election results, census data and their own membership Constituent Relationship Management (CRM) data. Plotted on a Google Map, they can identify interesting areas for them to canvass. This system facilitates efficient use of means:

We would do nothing more happily than knocking on every single door in every city, but unfortunately, we do not yet have that kind of manpower. So we do an analysis: What kinds of neighbourhoods are especially interesting for us? We have built our own system to help us make that decision [Campaign leader 11].

Organisational factors

Circumstances within the campaign’s organisation itself can form a barrier for the uptake of PBT-techniques. Less innovative parties, for instance, do not have a dedicated data, tech, or digital department. As a campaign leader notes: "The department responsible for that [tech/data/digital] is our Communication department. So that's four or five people. And sometimes someone of the department picks it up, but there's not one specific person who's responsible."This contrasts with GroenLinks, which has a Digital and Grassroots department and with SP's Digital department.

The "state of mind"within a campaign can also be seen as a barrier:"In the sense that internally, people are still very much inclined to think offline. The culture within the campaign is quite offline." [anonymous campaign leader]2

New leadership and younger staffers can play a facilitating role in political organisations. Campaign leader 11 argues that, because he is young, their new party chair brings a more tech-savvy vision than his predecessor. According to the campaign leader, younger staffers are more likely to implement tech and data in their work procedures.

A final organisational barrier is the primary goal a political party pursues. Campaign leader 10:

Maybe the strange thing about SGP is that we do not care that much about seat maximization. For us, it's about the impact of our principles. And sure, we would rather have four seats than three, but if we have to settle for three seats: that's fine too. And that's, in my opinion, a reason why we have a feeling like: do we really need data?

Structural electoral cycle factors

These circumstances are largely beyond the control of the campaigns, but they can influence the uptake of PBT-techniques. Campaign leaders see the PBT-actions of other political campaigns as a motivational factor. As campaign leader 11 notes about the development of their app: "I've looked a little bit at how GroenLinks have their app and canvassing system." Or as campaign leader 2 concludes: "If every party does it, you don't win very much by it. But if you're the only party that does nothing..."

Ethical and legal concerns

Especially D66 and the seniors’ party 50PLUS take a principled stance against the collection of data and the use of PBT. Where D66 presents itself as a privacy champion and therefore will never gather and use information about (groups of) voters, 50PLUS campaign leader 6 warns about the risk of irresponsible use of the data gathered by the "almost stalking of people", which he calls "morally irresponsible".

Furthermore, a lack of internal legal expertise appears to contribute to a feeling of legal uncertainty, which affects the likelihood of adopting PBT-techniques: "Legislation has grown so very comprehensive and complex. It's almost impossible to cope for us as a small organisation." [Campaign leader 5]

While ethical and legal concerns can form a barrier, a left- or right-wing orientation does not seem to be instrumental therein. After all, we have seen left-wing parties GroenLinks and SP develop relatively advanced PBT-tools. And we have seen right-of-centre party CDA express clear interest in advancing their own PBT capabilities. At the same time, left-of-centre 50PLUS and right-of-centre D66 both oppose the use of PBT.

System level

Electoral system

Although the Dutch one district PR-system should make for a rather equal distribution of campaign efforts, campaigns still divide the country into smaller areas of interest called 'key areas'. These areas differ per party, but do receive a relatively large part of campaign attention. Campaign leader 1 describes these as areas: "where we know the turnout is low, but the number of PvdA-voters is high". All campaigns use data provided by the Electoral Council, showing the election results per party, per voting location to establish key areas. Campaign leader 3 explains:

Using that [the election results], you see: Okay, we do well in this neighbourhood or this street. And then you combine that information with the CBS3 data, to find out what kind of neighbourhood it is, what kind of people live there, what are their backgrounds, how much do they earn, what does the family composition look like, et cetera.

Facilitated by these public data, campaigns enrich their knowledge of specific areas. A next step would be to use those data to make personalised appeals to (subgroups of) people living in those specific key areas.

Regulatory framework

Although the Netherlands would qualify as a minimally regulated environment (Plasser & Plasser, 2002; Esser & Strömbäck, 2012), campaigns all experience regulatory pressure and legal uncertainty on a system level. They cite an abundance of regulations, forming a barrier to their ability to innovate.

The technological developments have been taking place so very quickly. And, in that timeframe, to adjust all your procedures and everything. And also to meet the privacy regulations, I think many parties face a huge challenge in that respect." [Campaign leader 4]

Campaigns sometimes face a dilemma, having to decide between innovative techniques and privacy regulations. Campaign leader 11:

Regulations sometimes are unclear, which leads us to decide to go for the safe option because you do not know where the red line is. And you never want to abuse someone's personal data. So yes, regulations sometimes cause us to hit the brake and that's a good thing.

Culture

There is a recurring worry about the perceived low level of political knowledge of the average Dutch voter. PBT-techniques can facilitate campaigns' efforts to convince or educate such low-information voters, for example by "having a conversation with someone, especially if you share some characteristics," [Campaign leader 7] or by interesting "people for things that are relevant to them and to make them aware of the political dimension of those things." [Campaign leader 8] Campaign leader 3, in contrast, concludes that the electorate's low level of political knowledge (together with the perceived volatility of the electorate, the decrease in political trust, and their focus on persons instead of parties) forms an insurmountable barrier, making PBT-techniques irrelevant.

Discussion and conclusion

In the 2017 elections, used here as a case study, all campaigns use PBT through Facebook, but some parties are more advanced than others, and have even developed their own PBT-tools. We have established what the main barriers and facilitators for PBT are, using five factors on a campaign level and three factors on a system level. Not only does this study shed light on the conditions under which these barriers and facilitators manifest themselves, it also gives insight into their different workings across parties. Our study provides information about the data collected by parties and the PBT-techniques used to attract voters. We demonstrate how personal networks and cognitive diversity within a campaign can level barriers. We show how PBT is not only perceived as useful for campaigns in a FPTP-system, but in a PR-system as well. And we show how regulatory pressure is perceived as an obstacle and as a welcome 'normative red line'.

A triangulated research approach can improve our understanding of the campaign leaders' constructs. Observation of their (use of) PBT-tools and how these tools help campaigns make strategic decisions, can give more insight into the workings of these techniques. Another approach would be to interview canvassers and identify 'field-level' barriers and facilitators. Furthermore, ideally, we would have spoken to all parties holding a seat in parliament. Unfortunately, three parties did not cooperate. Two of those became the largest (VVD) and second largest (PVV; in a very close field) party. Since we did have access to eight of 11 parties, we are confident about our findings and we do not expect to identify additional factors influencing the adoption of PBT from interviews with the remaining parties.

Compared to related recent studies by Anstead (2017), Hersh (2015), Kreiss (2016), and Nielsen (2012), this study makes a number of contributions. In general, we focus our exploratory research on a PR-system instead of a FPTP-system, and we develop a model that takes system level contextual factors into account. Specifically, unlike Anstead (2017) we have found evidence for equalisation (which occurs when smaller parties take advantage of the internet's low costs and direct communication possibilities, and, in doing so, use the internet as a tool of empowerment [e.g. Margolis, Resnick, & Levy, 2003]). This evidence is especially clear in the case of GroenLinks, which was, at the time of the campaign, one of the smallest parties in parliament (now the fifth party). Furthermore, we provide an insightful point of view into Anstead's question of whether "parties develop data-driven capabilities more rapidly in electoral systems with a tendency towards disproportionate outcomes" (2017; p. 23). In comparison with Hersh (2015), we focus less on how differences in data-availability lead to different strategic decisions, and more on how differences in the perception of campaign level and system level factors lead to a variation in the occurrence in PBT-innovation. With regard to Kreiss (2016), we have extended his model and applied it to a multiparty democracy. In comparison with Nielsen (2012), we focus solely on the perception of campaign leaders and not on canvassers. Furthermore, we focus on PBT on online as well as offline platforms.

Our attention for the system level factors has enabled us to identify perceived influence of the PR-system on the adoption of PBT. Contrary to theoretical expectations (Plasser & Plasser, 2002), campaigns in a one-district PR-system do identify key-areas that are more heavily campaigned than other districts. These key-areas differ from 'battleground states' in FPTP-systems in the sense that the key-area does not sometimes swing one way and sometimes the other, but rather that potential voters in key-areas are supportive of a certain party, but not very likely to show up at the polls. Campaigns use PBT-techniques to convince these potential voters of the personal relevance of politics and to motivate them to cast their vote. Areas with firm turnout numbers and clear support for a certain party, in contrast, are perceived as less decisive and less of a priority. This leads to a hierarchy of areas, which differs per party. Also, as a PR-system typically leads to a relatively large number of parties partaking in an election, PBT can be seen as an asset for a campaign to organise in a more efficient manner. Moreover, according to the campaign leaders, PBT-techniques offered by Facebook do allow smaller parties a degree of visibility that they are unable to achieve through traditional media.

On a campaign level, in the coming years, we expect more citizen-initiated campaigning (Gibson, 2015) by campaigns low in labour-resources. This requires a solid infrastructure, which opens the door for third party intermediaries offering off-the-shelf infrastructure. In this regard, it would be interesting to track the development of PvdA, which has suffered its biggest loss in history. This critical event could lead to the prototyping (Kreiss, 2016) of GroenLinks' innovative campaign by PvdA. As the party's chairman has resigned, the door is open to a more cognitive diverse party structure (Du Pre, 2017; De Vaan et al., 2015). Of course, these developments might apply less to parties that are officially more cognisant of campaign ethics (e.g. D66). This is why ethics and legal aspects are important factors to take into consideration. It would be interesting to see how these campaigns act as PBT-capabilities of rival parties improve. Their self-imposed barrier can limit their future chances, but can also attract voters growing more aware of the value of privacy. In the former case, this could lead to an overhaul of their privacy principles, or perhaps to a legislative push towards the restriction of PBT (similar to Hersh, 2015). In the latter case, campaigns can be expected to develop innovative non privacy-invasive campaigning techniques. Either way, our model would provide tools to study the process.

So 'can political parties campaign in Europe as they do in North America' (Bennett, 2016)? We would say ‘mostly yes’. We agree with Bennett (2015) that there are important differences between the US and Europe, and indeed, they influence how PBT is used. But based on our findings, we are hesitant to conclude that those differences (severely) constrain the export of PBT-practices to European multiparty systems. We have shown that relatively small campaign budgets do not need to bar parties from engaging in PBT-practices (or even from cooperating with BSD, an 'expensive' American political consultancy). The same is true of the electoral system: campaign leaders generally perceive PBT-techniques as useful in a PR-system. What remains is the relatively strict Dutch data protection law, labelling political preference as 'sensitive personal data', which can only be processed with explicit consent from the potential voter. 'Explicit consent', however, sounds harsher on paper than it is in practice and is easily achieved (e.g., Beales & Muris, 2008; Calo, 2012; Joergensen, 2014). Of course, because of data regulations and/or their non-existence, European campaigns are unable to consult voting lists showing whether an individual showed up at the polls in the last elections. In most European countries, the electoral register is inaccessible to political parties. One might argue that, from a campaign's perspective, US voter data are superior to European voter data. We would argue that European data are different, but they do not bar European campaigns in the use of PBT-techniques. Dutch campaigns, for instance, can (and do) rely on election results on voting booth level (which comprises a couple of streets). They can (and do) combine these results with detailed, accurate, and a multitude of data about the neighbourhoods surrounding those voting booths. And then there is Facebook, facilitating easy targeting of its users with personalised messages. As potential challenges for democracy come with PBT, such as ignoring 'less valuable' citizens (e.g. reliable non-voters), more research into the workings and effects of PBT is needed.

Disclaimer: This guest essay in the Special issue on political micro-targeting has not been peer reviewed. It is treated here as a reflection.

During the past three decades there has been a persistent, and dark, narrative about political micro-targeting. Phil Howard (2006) vividly described a present and future where politicians would use data to “redline” the citizens that received political information, manufacturing attitudes and beliefs, leading to “managed citizenship”. In the years since Howard wrote his monumental book, the concerns over micro-targeting have only grown. The explosion of data about the electorate in Western democracies such as Australia, Canada, the UK, and the United States (Howard & Kreiss, 2010) has triggered deep unease among scholars and privacy advocates alike. Sophisticated voter databases now contain everything from political party data gleaned through millions of interactions with the electorate, public data obtained from state agencies, and commercial marketing information that is bought and sold on international open markets. The 2016 US presidential election revealed the new ways that individuals can be profiled, identified, found, tracked, and messaged to on social media platforms such as Facebook and YouTube, which these companies themselves help facilitate (Kreiss and McGregor, 2017).

While it might seem that the micro-targeting practices of campaigns have massive, and un-democratic, electoral effects, decades of work in political communication should give us pause. Although we lack the first-hand data from political campaigns, consultancies, and technology firms such as Facebook to know for sure, previous research tells us that people are seldom the unwitting dupes of strategic political communication. Partisanship shapes much of how people vote and decades of research reveals that it is really hard to change people’s minds through campaigns (Kalla & Broockman, 2017; Henderson & Theodoridis, 2017). This has large implications for the effectiveness of micro-targeting. For example, Eitan Hersh’s (2015) deeply and carefully researched ground-breaking study using data from a major vendor to the US Democratic Party discovers that campaign practitioners find it really hard to persuade voters. This is because political practitioners lack reliable and identifiable data on cross-pressured and low information voters. Given this, campaigns often focus on known voters rather than risk targeting and messaging to the wrong people. Indeed, Hersh reveals that despite hundreds of data points on members of the electorate, it is a small cluster of publically available data – such as turnout history, party identification, and demographic data – that matters far more for predicting vote choice.

The lesson is that micro-targeted campaign ads are likely most effective in the short run when campaigns use them to mobilise identified supporters or partisans, spurring volunteerism, donations, and ultimately turnout – hardly the image of a managed, manipulated, or duped public (Baldwin-Philippi, 2017). Ironically, campaigns often use micro-targeting to further these forms of democratic participation, making appeals to targeted subsets of voters on the basis of the parties and issues they already care about. Campaigns also use micro-targeting in the attempt to decrease voter turnout on the opposing side, sending negative messages to the oppositions’ likely voters in the hopes this will make them less excited to turn out for their candidate. But two decades of social science suggests that this can be a risky strategy given that partisans can rally behind their candidate who is being attacked (Dunn & Tedesco, 2017).

What explains the outsized concerns about micro-targeting in the face of the generally thin evidence of its widespread and pernicious effects? This essay argues that we have anxieties about micro-targeting because we have anxieties about democracy itself. Or, to put it differently, that scholars often hold up an idealised vision of democracy as the standard upon which to judge all political communication. In a world where many scholars and journalists both hope and ardently believe, in the face of all available evidence, that members of the public are fundamentally rational, seek to be informed, and consider the general interest, micro-targeting appears to be manipulative, perverting the capacity of citizens to reason about politics. Meanwhile, for many scholars and journalists, political elites are fundamentally opposed to members of the public, seeking domination or control as opposed to representing their interests. In this world, much of the concern over micro-targeting reads as a classic “third-person effect”, where scholars and journalists presume that members of the public are more affected by campaign advertising than they themselves are.

And yet, this idealised version is not how democracy really is, nor necessarily how it should be. The argument of this brief essay is that, as a quantifiable practice premised on strategically identifying targeted groups of voters and crafting messages designed to appeal to them, micro-targeting is broadly reflective of the fact that democracy is often partisan, identity-based, and agonistic – in short, political. Following communication scholar Michael Schudson’s (1986) study of commercial advertising nearly three decades ago, this essay asks the following questions in the US context: what is the work that micro-targeting does, where does it fit into the political culture, and, what kind of political culture has given rise to it? I argue that micro-targeting is only imaginable, and efficacious, in a polity that prizes partisan mobilisation, group solidarity, agonism, and the clash of opposing moral views in its politics. Following from this, I suggest different democratic concerns about micro-targeting that relate to its cultural power to, over time, create a powerful set of representations of democracy that undermines the legitimacy of political representation, pluralism, and political leadership.

The cultural work of micro-targeting

To analyse the role that micro-targeting plays in politics, first we need to understand how and why citizens vote. In their recent book Democracy for Realists, political scientists Christopher Achen and Larry Bartels (2016) offer a sustained critique of what they call the “folk theory” of American democracy. According to this “folk theory” that underlies conceptions of popular sovereignty, Americans have identifiable and consistent policy preferences. During the course of an election, they inform themselves about the policy positions of candidates and make rational decisions as to which best represents their preferences, which in turn leads parties to be responsive to the wishes of the public.

As Achen and Bartels (ibid.) argue, this is a fiction. They outline a “group theory of democracy”, where it is social attachments and group identification that largely determine both partisanship and vote choice. Achen and Bartels argue that people see themselves in relation to the groups that they belong to and those that they do not. Identity is so strong, in this account, that it conditions both what partisans believe parties stand for but also their interpretation of facts (ibid., 267; see also Prasad et al., 2009). As Achen and Bartels demonstrate, this identity and group theory of politics has expansive empirical support over seventy years of research which demonstrates, time and again, that people have little knowledge about politics and yet detailed understandings of the social groups that the Democratic and Republican parties are perceived to represent. It is in this context that candidate performances of partisan and social identity become more important for electoral outcomes then the informational content of journalism. Events and candidates make identity more or less salient and strengthen group attachments. During campaigns, parties and candidates work to remind voters of their partisan and social attachments and strengthen them so they are mobilised to participate in the election. As Achen and Bartels (ibid., 311) argue:

Political campaigns consist in large part of reminding voters of their partisan identities – “mobilizing” them to support their group at the polls. Formal communications by the groups and informal communication networks among group members also help citizens understand how their identity groups connect to the candidates and parties.

In this context, what is important about political campaigns is this work of communicating the partisan and social identities of candidates to voters. Candidates and their campaigns use micro-targeting, along with other strategic communications, to accomplish this. Micro-targeting is both a campaign practice of using data to craft and deliver strategic messages to subsets of the electorate (historically across many different media), and a genre of campaign communications that, much like political advertising more broadly, reinforces and amplifies the partisan, group, and identity conflicts at the heart of US politics. There has been extensive research on how micro-targeting works as a data-driven and quantifiable practice (see, for instance, Karpf, 2016). What these messages do as a genre of campaign communications, however, has received considerably less scrutiny. Drawing on my own previous work in the US context (Kreiss, 2016), the first argument that I develop here is that micro-targeting furthers the mobilisation that Achen and Bartels (2015) identify, primarily through reminding citizens of and shoring up their partisan and group identities. I then discuss the potential democratic consequences of this in a more expansive, cultural sense.

Micro-targeted ads have an aesthetic of what I call “political realism”, building on Michael Schudson’s work on commercial advertising. In Advertising, The Uneasy Persuasion, Schudson (1986) compared commercial advertising with Soviet realist art (the official state-sanctioned art of the former Soviet Union), arguing that it offers a form of “commercial realism”. As commercial realism, commercial advertising “simplifies and typifies” (215); advertising is abstracted, presenting the world as it should be, not as it is, and it exemplifies individuals as members of larger social groups. As it does so, “the aesthetic of capitalist realism — without a masterplan of purposes — glorifies the pleasures and freedoms of consumer choice in defense of the virtues of private life and material ambitions.” (ibid., 218) 

We can see micro-targeted digital advertising as a cultural form of ‘political realism’ that reflects, reinforces, and celebrates a political culture, at least in the United States, premised on identity, moral certainty, and mobilisation - not weighty considerations of the general interest or deliberation. Micro-targeted digital content shares a few central characteristics, which I adapt here for politics from Schudson’s (1986) work on commercial realism:

• It presents social and political life in simplified and typified ways;
• It presents life as it should become, or for negative ads, as it must not become;
• It presents reality in its larger social significance, not in its actual workings;
• It presents progress towards the future and positive social struggle, or for negative ads, the ideas of the other party as negative steps back into the past. It carries a message of optimism for one partisan side, and takes a stance of pessimism towards political opponents; and,
• It tells us that political conflict is necessary, a clash of different groups and worldviews; moral certainty is assured, political identity is certain, and political agonism is reality.

For example, micro-targeted ads present social life in simplified ways, not presenting actual lives but abstract, stylised ones designed to be rife with larger meaning. A depiction of a farmer’s daily work in a campaign ad, for instance, is not about actual events or daily labours, but is meant to be an abstract, simplified, symbol of the American values of hard work and cultivation of the earth and celebration of ordinary people in a democratic society. The farmer here is typified; the campaign ad is not about a real person who farms. The farmer is a representation of the larger social categories, values, and ideas the ad presents as desirable or worthy of emulation for all Americans. At the same time, the two dominant US political parties often stress different themes in their ads, a recognition that they have different visions of what life should be become, what progress is, and what worldviews and moral claims the public should embrace. While doing so, political micro-targeting is inherently pluralist. It reflects a basic claim that “everyone has interests to defend and opinions to advance about his or her own good, or the group’s good, or the public good, and every interest was at least potentially a political interest group.” (Rosenblum, 2010, 259)

While it is impossible to know the full range of micro-targeted ads run during the course of an election cycle, consider some of the examples culled from the non-profit and non-partisan Democracy in Action website that chronicles US campaigns and the Hillary for America Design 2016 website that compiles the creative design from the campaign. To start, much of political micro-targeting is about building campaign databases by finding supporters online, signing them up for the cause through email, and repeatedly messaging them to enlist them in becoming a volunteer or a donor.

Take, for instance, the declarative “I am a Hillary Voter” digital ad (see Figure 1), presumably (but also logically) directed (although we can never know for sure) at the candidate’s supporters. What separates micro-targeted political ads from their mass broadcast counterparts is the data that lies behind them: campaigns can explicitly try to find and send messages to their partisan audiences or intra-party supporters, linking the names in their databases to identities online or on social media platforms such as Facebook. Campaigns can also try to find additional partisans and supporters by starting with the online behaviours, lifestyles, or likes or dislikes of known audiences and then seeking out ‘look-alike audiences’, to use industry parlance. And, what people do when they see these ads is quantified in terms of their performance, measured through things such as engagement and click-throughs. Micro-targeting is about mobilisation through conveying and building social solidarity. While there is much concern over candidates speaking out of both sides of their mouths to the electorate through hyper-targeted digital ads, likely far more often campaigns use micro-targeting to provide occasions for social identification and group belonging, conveying and constructing the sense of shared identity and group membership at the heart of politics. The “Wish Hillary a Happy Birthday” ad captures this (see Figure 2). Not only is this appeal directed at supporters (what Republican will want to wish Hillary a happy birthday after all), it constructs a sense of what social identification with Hillary Clinton means: motherhood, family, warmth, care, and nurturing.

Figure 1: Hillary Clinton digital campaign advertisements

Figure 2: Hillary Clinton digital campaign advertisement

Source: Hillary for America Design 2016

Micro-targeting is also about the marking of difference. This is, perhaps, the most common trope in micro-targeted digital campaign ads. Campaigns look to not only establish the cultural meaning of their candidates and supporters, but also that of their opposition (Alexander, 2010). Donald Trump’s ads during the 2016 election reflected his rhetoric from the campaign trail in stressing themes of safety and security, in addition to the need to draw boundaries around civic incorporation (i.e., who should be allowed to be a citizen). For Hillary Clinton, micro-targeted ads were celebrations of diversity and multi-culturalism, especially the empowerment of women and racial and ethnic minorities. Political advertisements attempt to connect the candidates they promote with the demographic and social groups they seek to represent (in the United States this is at times drawn on racial and ethnic terms: whites for Republicans and a more diverse coalition for Democrats, see the discussion in Grossmann & Hopkins, 2016, 43-45).

In this, micro-targeting reflects and reinforces political agonism, the clash of competing social groups, interests, and values. Through micro-targeting, candidates stake out their claim to be on the civil side of the moral binary of the political sphere and strive to paint their opponents as anti-civil (Alexander, 2010). More colloquially, micro-targeted advertisements offer the beautiful affirmation of our values and the sharp critique of those of our opponents. Hillary Clinton’s campaign, for instance, clearly sought to portray Trump in terms of anti-civil racism, xenophobia, and sexism. And, the campaign used issues, such as abortion rights, and values, such as autonomy and choice, to build group identity and social solidarity around opposition to Trump: “Let’s stand together, join millions of women” (see Figure 3). This Facebook ad pits Clinton and her supporters against Trump and his supporters. Trump, in turn, combined nationalist and security appeals with an implicit construction of the American body politic in white identity terms (Figure 4). These ads capture the reality that political conflict is not only inevitable, but necessary: there are opposing views in politics on fundamental questions such as life, autonomy, and country. The audiences for these ads are not being presented with information to help them make up their own minds, they are being invited into a political struggle with clear opposing worldviews and moral values (see Figure 5). This is why mobilisation ads are directed towards identity-congruent audiences.

Figure 3: Hillary Clinton Facebook advertisement

Figure 4: Donald Trump digital advertisement

Source: Democracy in Action

Figure 5: Anti-Trump Hillary Clinton digital advertisements

Source: Hillary for America Design 2016

In these advertisements, it is also clear that micro-targeted ads present life as it should become, or as it must not become, linking the preferred candidate and political party with a civil vision of the future and the opposition with an anti-civil vision of the future, to use Alexander’s (2010) framework. As an example, for Ted Cruz (see Figure 6), the opposing side wants to infringe on the Bill of Rights, the fundamental liberty of Americans to defend their lives, liberties, families, and properties. Candidates run these issue ads to stake out their stance on the conflicting values, visions of the good life, plans for the future, and ends that are desirable in politics – whether it is embracing the freedom and security of gun rights for American Republicans or autonomy and choice in the context of reproductive rights for Democrats. These appeals are designed to mobilise the committed around the candidate’s vision of America’s past and future – they are designed for a world where we are sure of who we are and committed to our values and the ends we pursue.

Figure 6: Ted Cruz digital campaign advertisement

Source: Democracy in Action

Conclusion: democratic anxieties

I believe that there is such democratic anxiety about micro-targeting because citizens are supposed to be independent, autonomous, and rational. Micro-targeted advertising works to reinforce group identities and solidarity, mobilise partisans, and further the clash of political values. These things are all suspect from the perspective of the powerful and potent “folk theory” of democracy, as Achen and Bartels phrase it. As these realists argue, however, it’s far better to grapple with the reality of group-based democracy, with its attendant ingrained social allegiances and conflicts over values and power, rather than wishing for a transcendent and pure form of democracy without politics. These authors argue that we need to make peace with conflictual and competitive forms of group-based and pluralistic democracy premised on institutionally organised opposition. As Achen and Bartels (2015, 318) conclude:

Freedom is to faction what air is to fire, Madison said. But ordinary citizens often dislike the conflict and bickering that comes with freedom. They wish their elected officials would just do the people’s work without so much squabbling amongst themselves. They dislike the compromises that result when many different groups are free to propose alternative policies, leaving politicians to adjust their differences. Voters want “a real leader, not a politician,” by which they generally mean that their own ideas should be adopted and other people’s opinions disregarded, because views different from their own are obviously self-interested and erroneous. To the contrary, politicians with vision who are also skilled at creative compromise are the soul of successful democracy, and they exemplify real leadership.

My own view is that micro-targeting comes in the necessary service of this “conflict and bickering”. At its normative best, micro-targeting strengthens the hands of opposing factions, enabling them to identify and mobilise partisans to their cause, providing them with resources in terms of boots on the ground and money in the coffers. When opposing politicians and parties square off, they carry these resources into battle trying to advance their agendas or win concessions for their side. Compromise may be harder in a world of stronger factions, their hands steadied by the resources that micro-targeting can deliver, but that does not make compromise any less necessary or essential.

On the other hand, there are reasons for democratic concern about micro-targeting, but they look a bit different from narratives about public manipulation. Schudson (1986, 232) concludes that “advertising does not make people believe in capitalist institutions or even in consumer values, but so long as alternative articulations of values are relatively hard to locate in the culture, capitalist realist art will have some power.” I suspect that the same is true of political micro-targeting. The cultural power of political micro-targeting, but also political advertising more generally, lies in its creation of a set of ready-to-hand representations of democracy that citizens can express easily and fall back on. Taken to its extreme in a polarized political climate, micro-targeting can work to undermine the legitimacy of conflicts over opposing values and claims in democratic life. For example, in an undemocratic political culture micro-targeting can portray the other side as crooked and dangerous to the polity, political compromise as selling out, political expertise and representation as not to be trusted, and partisans’ own beliefs and identities as the only legitimate ones, not simply those among many in a pluralistic democracy. Micro-targeting also melds symbolic and social power in new ways, culturally legitimating and furthering the fortunes of autonomous and independent candidates, divorced from their parties and taking their appeals directly to voters (see Hersh, 2017).

Disclaimer: This guest essay in the Special issue on political micro-targeting has not been peer reviewed. It is an abbreviated version of a speech delivered by the Member of the European Partiament (MEP) Sophie in ‘t Veld in Amsterdam in May 2017 to Data & Democracy, a conference on political micro-targeting.

Democracy

Democracy is valuable and vulnerable, which is reason enough to remain alert for new developments that can undermine her. In recent months, we have seen enough examples of the growing impact of personal data in campaigns and elections. It is important and urgent for us to publicly debate this development. It is easy to see why we should take action against extremist propaganda of hatemongers aiming to recruit young people for violent acts. But we euphemistically speak of 'fake news' when lies, 'half-truths’, conspiracy theories, and sedition creepily poison public opinion.

The literal meaning of democracy is 'the power of the people'. 'Power' presupposes freedom. Freedom to choose and to decide. Freedom from coercion and pressure. Freedom from manipulation. 'Power' also presupposes knowledge. Knowledge of all facts, aspects, and options. And knowing how to balance them against each other. When freedom and knowledge are restricted, there can be no power.

In a democracy, every individual choice influences society as a whole. Therefore, the common interest is served with everyone's ability to make their choices in complete freedom, and with complete knowledge.

The interests of parties and political candidates who compete for citizen’s votes may differ from that higher interest. They want citizens to see their political advertising, and only theirs, not that of their competitors. Not only do parties and candidates compete for the voter's favour. They contend for his exclusive time and attention as well.

Political targeting

No laws dictate what kind of information a voter should rely on to be able to make the right consideration. For lamb chops, toothpaste, mortgages or cars, for example, it’s mandatory for producers to mention the origin and properties. This enables consumers to make a responsible decision. Providing false information is illegal. All ingredients, properties, and risks have to be mentioned on the label.

Political communication, however, is protected by freedom of speech. Political parties are allowed to use all kinds of sales tricks.

And, of course, campaigns do their utmost and continuously test the limits of the socially acceptable.

Nothing new, so far. There is no holding back in getting the voters to cast their vote on your party or your candidate. From temptation with attractive promises, to outright bribery. From applying pressure to straightforward intimidation.

Important therein is how and where you can reach the voter. In the old days it was easy: Catholics were told on Sundays in church that they had no other choice in the voting booth than the catholic choice. And no righteous Catholic dared to think about voting differently. At home, the father told the mother how to vote. The children received their political preference from home and from school. Catholics learned about current affairs via a catholic newspaper, and through the catholic radio broadcaster. In the Dutch society, which consisted of a few of such pillars, one was only offered the opinions of one's own pillar1. A kind of filter bubble avant la lettre.

Political micro-targeting

Nowadays, political parties have a different approach. With new technologies, the sky is the limit.

Increasingly advanced techniques allow the mapping of voter preferences, activities, and connections. Using endless amounts of personal data, any individual on earth can be reconstructed in detail. Not only can their personal beliefs be distilled from large troves of data, no, it even is possible to predict a person's beliefs, even before they have formed them themselves. And, subsequently, it is possible to subtly steer those beliefs, while leaving the person thinking they made their decision all by themselves.

As often is the case, the Americans lead in the use of new techniques. While we Europeans, touchingly old-fashioned knock on doors and hand out flyers at Saturday's market, the American employ the latest technology to identify, approach, and influence voters.

Of course, trying to find out where voters can be reached and how they can be influenced is no novelty. Political parties map which neighbourhoods predominantly vote for them, which neighbourhoods have potential, and in which neighbourhoods campaigning would be a wasted effort. Parties work with detailed profiles and target audiences, for which they can tailor their messages.

But the usage of personal data on a large scale has a lot more to offer. Obviously, this is a big opportunity for political parties, and for anyone else, who runs campaigns or aims to influence the elections.

However, the influencing techniques become increasingly opaque. As a result of the alleged filter bubble, voters are being reaffirmed in their own beliefs, and they hardly receive information anymore about the beliefs and arguments of other groups. This new kind of segmentation may stifle critical thinking. There may not be enough incentive to test one's own ideas, to find new arguments, or to critically reflect on the truthfulness of information.

I am a social and economic liberal D66 politician, and I get suggestions for news articles from websites like The Guardian or Le Monde. My colleague from the right wing nationalist PVV, may well receive URLs from Breitbart.

Pluralism is essential for a healthy, robust democracy. In a polarised society, people live in tightly knit groups, which hardly communicate with each other. In a pluralist society people engage in the free exchange, confrontation, and fusion of ideas.

The concept pluralism is under pressure. Populist parties declare themselves representative of The People. In their vision, The People, is uniform and homogenous. There is a dominant cultural norm, dictated from the top-down, to which everyone must conform. Whomever refuses, gets chewed out. Often, it is about one-dimensional symbolism such as Easter eggs and Christmas trees. There is no place for pluralism in the world of the populists. But when there is no pluralism, there is no democracy. Without pluralism, democracy is nothing more than a simple tribal dispute, instead of the expression of the will of all citizens together.

Voter data

European privacy legislation limits the use of personal data. In the world of ‘big data’, one of the explicit goals of regulation is to prevent restriction of the consumer's choice. Oddly enough, lawmakers do not explicitly aspire to guarantee voters as broad a choice as possible. But in politics, individual choices have consequences for society as a whole.

In 2018, the General Data Protection Regulation (GDPR) comes into effect. We have worked five years on the GDPR. At this moment, we work on the modernisation of the e-Privacy Directive, which is mainly about the protection of communication. As was the case with the GDPR, companies from certain sectors scream bloody murder. European privacy protection would mean certain death for the European industry. According to some corporate Cassandras, entire European industries will move to other continents. That very same death of corporate Europe is also predicted for any measure concerning, say, environmental norms, procurement rules, or employee rights. All those measures are in place, but, as far as I know, the nightmare scenario has never occurred...

There are some corporate sectors, such as publishing and marketing, which have a huge impact on the information supply to citizens. They are the ones who now cry wolf. It is understandable that they are unhappy with stricter rules concerning their activities, but as the potential impact of the use of personal data and ‘big data’ increases, so does their social responsibility.

At the moment, there is not much public debate about the new techniques. Peculiar. Thirty years ago, 'subliminal advertising', as we called it then, was prohibited because people found it unethical to influence people without their knowledge. We need to have a similar debate. What do we think of opaque influencing? Do we need ethical norms? Should such norms apply only to political campaigns, or should we look at this from a broader perspective? In the ‘big data’ debate, we tend to speak in technical or legal terms, while actually the issue is fundamentally ethical, holding far-reaching consequences for the vitality of our democracy.

Such a public debate demands more clarity on the impact of ‘big data’, profiling, targeting, and similar techniques on the individual, her behaviour, and her choices, which determine in what direction society progresses. Which voters are being reached? How sensitive are they for the subtle influencing and what makes them resilient? How do people who are hardly reached only compare to the others? How do voters and non-voters compare? Is the voter truly predictable? Can we identify or influence the floating voter? Do voters actually float between different parties? Or do they especially float within their own party, their own bubble, their own segment? How important are other factors, such as the social context? If the new influencing techniques are indeed as potent as we think, how can polls get it so wrong? What can we learn from advertisers who return to contextual advertising, because targeting turns out less effective than they thought?

We need to stay cool-headed. New technologies have a huge impact, but human nature will not suddenly change due to ‘big data’ and its use. Our natural instincts and reflexes will definitely not evolve in a few years. That would take many thousands of years, as even in the 21st century, we seem to have more than a few cavemen traits, so losing internalised behaviour is not as easy as 1-2-3. Humans are resilient, but democracy is vulnerable. On a short term, the societal impact is large. This gives us all the reason to reflect on how to deal with the new reality, and how we can keep up our values in this new reality.

The use of personal data, clearly, is not solely reserved for decent political parties. Other persons and organisations, from the Kremlin to Breitbart, can bombard European voters with information and misinformation. But European governments, controlling endless amounts of personal data of their citizens, can also manipulate information, or circulate utter nonsense to advance their own interests. A random example: the Hungarian government influencing their voters with lies and manipulation about the so-called consultation on asylum seekers.

Beyond voter data

This issue is not only about the personal data of voters, but also about the personal data of political competitors, opponents, and critics, which are increasingly being employed. Recently, we have seen efforts of external parties to influence the results of the 2017 French elections. We saw a large-scale hack of the Emmanuel Macron campaign, and the spread of false information, coming obviously from the Kremlin and the American Alt-Right, meant to discredit Macron's candidacy.

Also, the American elections show the shady game of hacking, leaking, and manipulating. The issue of the Hillary Clinton mails will undoubtedly occupy our minds for years. Who knows how the elections would have turned out without this affair?

Other democratic pillars can get corrupted as well by the misuse of data. Critical voices, opposition, and checks and balances are democracy's oxygen. Democracy is in acute jeopardy when data are employed to attack, undermine, discredit, blackmail, or persecute journalists, judges, lawyers, NGOs, whistleblowers, and opposition parties.

In Europe, we tend to shrug our shoulders at these dangers. "Oh well, we'll see, such things occur only in banana republics, not right here". Of course, this trust in our democratic rule of law is wonderful. But if we treat our rule of law this neglectfully, we will lose it eventually.

Within the European Union, we currently see this happening in Poland and Hungary. The governments of both nations ruthlessly attack independent judges, critical media, inconvenient NGOs. They do so with quasi-lawful means. Under the banner of transparency, they force NGOs to register. In doing so, they misuse laws against money laundering, and terror finance. Or the governments bring out compromising information about judges or politicians in strategic moments.

But critical voices struggle in other member states as well. Lawyers are being monitored, even without a legal basis. In the years after 9/11, we have created endless new abilities for intelligence services, police and justice departments to spy on citizens, even without suspicion, without the signature of a judge. The companies to which we unwittingly surrender our personal data, in exchange for service, are forced to hand over all information to the government, or forced to build in backdoors. Governments hack computers in other countries. Usually, it starts out with unlawful practices, but soon enough laws are put in place to legalise those practices. The magic word 'terrorism' silences any critique on such legislation.

But when politicians, journalists, NGOs, whistleblowers, lawyers, and many others cannot perform their tasks freely and without worry, our democracy withers. Not only do they have to operate without someone keeping an eye on them, they have to know nobody is in fact watching them. The mere possibility of being watched, results in a chilling effect.

For this principal reason, I have contested a French mass surveillance law before the French Conseil d'Etat. Since, as a member of the European Parliament, I spend four days a month on French soil (in Strasbourg), I could potentially be the target of the French eavesdropping programme. This is not totally imaginary, as I am not only a politician, but also a vocal critic of certain French anti-terror measures. It is not about me actually worrying about being spied on, but about the fact that I might be spied on. Luckily, I am not easily startled, but I can imagine that many politicians are vulnerable. That is a risk for democracy.

I do not discard the possibility of a ruling of the European Court of Human Rights on my case. In that turn of events, it will lead to jurisprudence valid in the entire EU (and the geographical area covered by the Council of Europe).

But, of course, this should not depend on the actions of one obstinate individual whether politicians, NGOs, journalists, and so on, can do their jobs fearlessly, to fulfil their watchdog role.

It is my personal, deep, conviction that the biggest threat to our democracy is the fact that we have enabled the powerful to access, with almost no limitations, the personal data of those who should control those very same powerful entities.

What can we do?

Some propose new forms of democracy, in which universal suffrage is weakened or even abolished. In his book ‘Against elections: the case for democracy’, David Van Reybrouck had the idea to appoint representatives on the basis of chance, and in his book ‘Against democracy’ Jason Brennan wants to give the elite more votes than the lower classes, presuming that people with more education or development make better choices. Others want to replace representative democracy with direct democracy.

I oppose those ideas. Universal suffrage and the representative democracy are great achievements, which have led to enormous progress in society.

First of all, we have to make sure our children grow up to be critical, independent thinkers. Think differently, deviate, provoke: this must be encouraged instead of condemned. A democracy needs non-conformists.

We must teach our children to contextualise information and to compare sources.

The counterpart of ‘big data’ must be ‘big transparency’. We need to understand not just open administration, but also insights into the techniques of influence.

The regulation and limitation of the use of personal data, as I hope to have argued effectively, is not a game of out-of-touch privacy activists. It is essential for democracy. We need safeguards, not only to be sure people really are free in their choices, but also to protect the necessary checks and balances. As such, I plea for a rigorous application of the GDPR, and in the European Parliament, I will work for a firm e-Privacy Directive.

And yes, perhaps we should examine whether the rules for political campaigning are still up-to-date. In most countries, those rules cover a cap on campaign expenditures, a prohibition of campaigning or polling on the day before election day, or a ban on publishing information that may influence the election results, such as the leaked e-mails in France. But these rules have little impact on the use of personal data to subtly influence elections.

Last year, the European Parliament supported my proposal for a mechanism to guard democracy, the rule of law, and fundamental rights in Europe.2

On this day (editor’s note: 9 May, Europe Day) of European democracy, I plead for equal, high norms in Europe. The last years have shown that national elections are European elections. It is crucial for us to trust that all elections in EU member states are open, free, and honest elections, free of improper influencing.

These last sixty years, the European Union has developed itself into a world leader in democracy and freedom. If we start a public debate, Europe can remain a world leader.

This paper is part of 'A Manchurian candidate or just a dark horse? Towards the next generation of political micro-targeting research’, a Special issue of the Internet Policy Review.

Introduction

Political campaigns in the United States have employed digital technologies for more than a decade, developing increasingly sophisticated tools and techniques during each election cycle, as “computational politics” have become standard operating procedure (Tufecki, 2014; Kreiss, 2016). However, the most recent election marked a critical turning point, as candidates, political action committees, and other interest groups were able to take advantage of significant breakthroughs in data-driven marketing techniques, such as cross-device targeting, developed since the previous presidential election (“Bernie Sanders”, 2016; Edelman Digital, 2016). Electoral politics has now become fully integrated into a growing, global commercial digital media and marketing ecosystem that has already transformed how corporations market their products and influence consumers (Chahal, 2013; LiveRamp, 2015; Rubinstein, 2014; Schuster, 2015).The strategies, technologies, and tools of digital political marketing are more complex and far-reaching than anything we have seen before, with further innovations already underway (WARC, 2017). But because most commercial and political digital operations take place below the radar, they are not fully understood by the public. 1

In the following pages, we briefly describe the growth and maturity of digital marketing, highlighting its basic features, key players, and major practices. We then document how data-driven digital marketing has moved into the centre of American political operations, along with a growing infrastructure of specialised firms, services, technologies and software systems. We identify the prevailing digital strategies, tactics, and techniques of today’s political operations, explaining how they were employed during the most recent US election cycle. Finally, we explore the implications of their use for democratic discourse and governance, discussing several recent policy developments aimed at increasing transparency and accountability in digital politics.

Our research for this paper draws from our extensive experience tracking the growth of digital marketing over the past two decades in the United States and abroad, monitoring and analysing key technological developments, major trends, practices and players, and assessing the impact of these systems in areas such as health, financial services, retail, and youth (Chester, 2007; Montgomery, 2007; Montgomery, Chester, & Kopp, 2017). During the 2016 US presidential election, we monitored commercial digital advertising and data use by candidates, parties and special interest groups across the political spectrum. We collected examples of these ads, along with technical and market impact information from the developers of the applications. We also reviewed trade journals, research reports, and other industry documents, and attended conferences that were focused on digital technologies and politics. In the process, we identified all of the major providers of political digital data targeting applications (e.g., Google, Facebook, data clouds, ad agencies) and analysed all their key materials and case studies related to their 2016 operations. The source for much of this work was our ongoing gathering and analysis of cross-sectional commercial digital marketing practices worldwide.

Marriage of politics and commerce

Since the mid-20^th century, advertising has been an increasingly powerful and pervasive presence in US political campaigns, as a growing cadre of ad agencies, public relations firms, and consultants perfected the use of opinion polls, focus groups, and psychographics to reach and influence voters through radio, television, direct mail, and other media outlets (A. Jamieson, 2016; K. H. Jamieson, 1996; Sabato, 1981). With the rise of the internet, campaign operatives began to harness digital technologies and tools to mobilize voter turnout, engage young people, raise money, and support grassroots ground operations (Karpf, 2016; Kreiss, 2016; Tufecki, 2014). Both major political parties in the United States developed large, sophisticated data and digital operations (Kreiss, 2016).

Many of the digital strategies, tools, and techniques employed in the 2016 election were initially developed, deployed, tested, and refined by the commercial sector (Tufecki, 2014).Since its origins in the mid-1990s, digital marketing has operated with a core business model that relies on continuous data collection and monitoring of individual online behaviour patterns (Montgomery, 2011). This system emerged in the United States amid a political culture of minimal government interference, and within a prevailing laissez-faire ethos regarding the internet and new technologies (Barlow, 1996). In the earliest days of the “dot-com boom”, a strong political alliance was forged between the digital media companies and their partners in the advertising and media business, enabling the nascent industry to effectively ward off any attempts to restrain its business operations through privacy regulation or other public policies (Solon & Siddiqui, 2017). As a consequence, the advertising industry played a central role in shaping the operations of platforms and applications in the digital media ecosystem. Digital marketing is now well established and thriving, with expenditures reaching nearly $72.5bn in 2016 for the US alone, and worldwide spending predicted to reach more than $223bn this year (eMarketer, 2017; IAB, n.d.-d).

Ongoing innovations over the years have increased the capacity of data and digital marketing applications. Data collection, analysis, and targeting were further woven into the daily lives of consumers with the rise of social media platforms and mobile devices. Because of the unique role that they play in users’ lives, these platforms are able to sweep up enormous amounts of information, including not only what users post about themselves, but also what is collected from them throughout their daily activities (Smith, 2014). A growing arsenal of software and analytic tools has enhanced the ability of digital media companies and their advertisers to glean valuable insights from the oceans of data they generate (Smith, 2014). Predictive analytics introduced an expanded set of tools for scoring, rating, and categorising individuals, based on an increasingly granular set of behavioural, demographic, and psychographic data (“What is Predictive Intelligence”, 2017). US digital marketers have helped popularise and spur the successful adoption of digital advertising platforms and applications in nearly every geographical location with an internet connection or a link to a mobile device (IAB, n.d.-c). Google, Facebook, and other major players in the digital marketing industry have also developed a global research infrastructure to allow them, and especially their major advertising clients, to make continuous improvements in reaching and influencing the public, and to measure with increasing accuracy the success of their efforts (Facebook IQ, n.d.-a). These developments have created what some observers have called the “surveillance economy” (Singer, 2012).

The growth of data-driven political marketing

Though political campaigns have employed micro-targeting techniques—which use an array of personalised and other data sets and marketing applications to influence the actions of individuals—during the last several election cycles, recent technological innovations and industry advances have created a much more robust system than what was in place in 2012 (IAB, n.d.-b; Rubinstein, 2014). For years, political campaigns have been able to combine public voter files with commercial information from data brokers, to develop detailed and comprehensive dossiers on American voters (Rubinstein, 2014). With recent advances in the advertising technology and data industries, they can now take advantage of a growing infrastructure of specialty firms offering more extensive resources for data mining and targeting voters. Among the new entities are data marketing clouds. Developed by well-known companies such as Adobe, Oracle, Salesforce, Nielsen, and IBM, these clouds sell political data along with an exhaustive amount of detailed consumer information for each potential target, including, for example, credit card use, personal interests, consumption patterns, and TV viewing patterns (Salesforce DMP, 2017).

Some of these massive cloud services also operate what has become a new and essential component for contemporary digital targeting—the data management platform (DMP) (Chavez, 2017). DMPs provide marketers with “centralized control of all of their audience and campaign data” (BlueKai, 2011). They do this by collecting and analysing data about individuals from a wide variety of online and offline sources, including first-party data from a customer’s own record, such as the use of a supermarket loyalty card, or their activities captured on a website, mobile phone, or wearable device; second-party data, information collected about a person by another company, such as an online publisher, and sold to others; and third-party data drawn from thousands of sources, comprising demographic, financial, and other data-broker information, including race, ethnicity, and presence of children (O’Hara, 2016). All of this information can be matched to create highly granular “target audience segments” and to identify and “activate” individuals “across third party ad networks and exchanges”. DMPs are quickly becoming a critical tool for political campaigns (Bennett, 2016; Kaye, 2016, July; Regan, J., 2016).

Facebook and Google now play a central role in political operations, offering a full spectrum of commercial digital marketing tools and techniques, along with specialied ad “products” designed for political use (Bond, 2017). Not surprisingly, these companies have also made generating revenues from political campaigns an important “vertical” category within their ad business (Facebook, n.d.-d; Facebook IQ, n.d.-b; Stanford, 2016). Facebook’s role in the 2016 election was particularly important. With users required to give their real names when they sign up as members, Facebook has created a powerful “identity-based” targeting paradigm, enabling political campaigns to access its more than 162 million US users and to target them individually by age, gender, congressional district, and interests (Facebook, n.d.-b).Its online guide for political campaign marketing urges political campaigns to use all the social media platform tools it makes available to advertisers—including through Instagram and other properties—in order to track individuals, capture their data through various “lead-generation” tactics, and target them by uploading voter files and other data (Facebook, n.d.-a-c-f). The company also employs teams of internal staff aligned with each of the major political parties to provide technical assistance and other services to candidates and their campaigns (Chester, 2017; Kreiss & Mcgregor, 2017). Google heavily promoted the use of YouTube, as well as its other digital marketing assets, during the 2016 US election, reaching out to both major political parties (YouTube, 2017).

The growth and increasing sophistication of the digital marketplace has enhanced the capacities of political campaigns to identify, reach, and interact with individual voters. Below we identify seven key techniques that are emblematic of this new digital political marketing system, providing brief illustrations of how they were employed during the 2016 election.

Cross-device targeting

Getting a complete picture of a person’s persistent “identity” through an “identity-graph” has become a key strategy for successfully reaching consumers across their “omnichannel” experience (use of mobile, TV, streaming devices, etc.) (Winterberry Group, 2016). “Cross-device recognition” allows marketers to determine if the same person who is on a social network is also using a personal computer and later watching video on a mobile phone. Through data “onboarding,” a customer record that may contain a physical and email address is linked through various matching processes, associating it with what is believed to be that individual’s online identification—cookies, IP addresses, and other persistent identifiers (Levine, 2016). Cross-device targeting is now a standard procedure for political initiatives and other campaigns. Voter files are uploaded into the onboarding process, enabling the campaigns to find their targets on mobile devices and at specific times when they may be more receptive to a message (Kaye, 2016, April; L2, n.d.-b). Such granularity of information also enables a more tailored advertisement—so-called “dynamic creative”—which can be changed over time to “deliver very specific messaging” to individuals (Schuster, 2015). Leading cross-device marketing company Drawbridge offered a suite of election services in 2016 that provided campaigns a number of ways to impact voters, including through “Voter-Centric Cross Device Storytelling”, “Political Influencer Identification”, and via “Real-Time Voter Attribution Measurement” (Drawbridge, n.d.).

Programmatic advertising

Programmatic advertising refers to new automated forms of ad buying and placement on digital media using computer programmes and algorithmic processes to find and target a customer wherever she goes. The process can also involve real-time “auctions” that occur in milliseconds in order to “show an ad to a specific customer, in a specific context” (Allen, 2016). The use of programmatic advertising was one of the major changes in political campaign digital operations between 2012 and 2016—“the first time in American History”, according to one ad company, “that such precise targeting has ever been made available at such great scale” (Briscoe, 2017; Kaye, 2015). Programmatic advertising has itself grown in its capabilities to reach individuals, taking advantage of new sources of data to reach them on all of their devices (Regan, T., 2016). In 2016, for example, global ad giant WPP’s Xaxis system—“the world’s largest programmatic and technology platform”—launched “Xaxis Politics”. Capable of “reaching US voters across all digital channels”. the system is said to “segment audiences by hundreds of hot button issues as well as by party affiliation”, including via “real-time campaigns tied to specific real-world events” (Xaxis, 2015). Candidates were able to use the services of a growing list of companies, including Google, Rubicon, AOL, PubMatic, Appnexus and Criteo, that offered programmatic advertising platforms (“Political Campaigns”, 2016; Yatrakis, 2016).

Lookalike modelling

The use of big data analytics enables marketers to acquire information about an individual without directly observing behaviour or obtaining consent. They do this by “cloning” their “most valuable customers”in order to identify and target other prospective individuals for marketing purposes (LiveRamp, 2015). For example, Stirista (n.d.), a digital marketing firm that also serves the political world, offers lookalike modelling to identify people who are potential supporters and voters. The company claims it has matched 155 million voters to their “email addresses, online cookies, and social handles”, as well as “culture, religion, interests, political positions and hundreds of other data points to create rich, detailed voter profiles”. Facebook offers a range of lookalike modelling tools through its “Lookalike Audiences” ad platform. For example, Brad Parscale, the Trump campaign’s digital director, used the Lookalike Audiences ad tool to “expand” the number of people the campaign could target (Green & Issenberg, 2016). Facebook’s “Custom Audiences” product, similarly, enables marketers to upload their own data files so they can be matched and then targeted to Facebook users (Facebook, n.d.-e).

Geolocation targeting

Mobile devices continually send signals that enable advertisers (and others) to take advantage of an individual’s location—through the phone’s GPS (global positioning system), Wi-Fi, and Bluetooth communications. All of this can be done with increasing speed and efficiency. Through a host of new location-targeting technologies, consumers can now be identified and targeted wherever they go, while driving a car, pulling into a mall, or shopping in a store (Son, Kim, & Shmatikov, 2016). A complex and growing infrastructure of geolocation-based data-marketing services has emerged, with specialised mobile data firms, machine-learning technologies, measurement companies, and new technical standards to facilitate on-the-go targeting (Warrington, 2015). The use of mobile geo-targeting techniques played a central role in the 2016 election cycle, with a growing number of specialists offering their services to campaign operatives. For example, L2 (n.d.-a) made its voter file, along with HaystaqDNA modelling data, available for mobile device targeting, offering granular profile data on voters based on their interest in such contested topics as gun laws, gay marriage, voter fraud, and school choice, among others. Conde Nast’s Advance Publications’ election advertising worked with campaigns to append geo-location, profile data, and buying behaviour “to sculpt a very specific voter profile and target down to few hundred readers in a given geo location” (Ellwanger, 2016).

Online video advertising

Digital video, via mobile and other devices, is perceived as a highly effective way of delivering emotional content on behalf of brands and marketing campaigns (IAB, n.d.-a). There are a variety of online video ad formats that provide both short- and long-form content, and that work well for political and other marketing efforts. Progressive political campaign company Revolution Messaging, which worked for the Sanders campaign, developed what it calls “smart cookies” that it says take video and other ad placement “to the next level, delivering precision and accuracy” (Revolution Messaging, n.d.). Google’s YouTube has become a key platform for political ads, with the company claiming that today, voters make their political decisions not in “living rooms” in front of a television but in what it calls “micromoments” as people watch mobile video (DoubleClick, n.d.). According to the company’s political ad services research, mobile devices were used in nearly 60 percent of election-related searches during 2016. Content producers (which it calls “Creators”) on YouTube were able to seize on these election micro-moments to influence the political opinions of potential voters 18-49 (“Letter from the Guest Editors,” 2016).

Targeted TV advertising

Television advertising, which remains a linchpin of political campaign strategy, is undergoing a major transformation, as digital technologies and “addressable” set-top boxes have changed cable and broadcast TV into powerful micro-targeting machines capable of delivering the same kinds of granular, personalised advertising messages to individual voters that have become the hallmark of online marketing. Political campaigns are in the forefront of using set-top box “second-to-second viewing data”, amplified with other data sources, such as “demographic and cross-platform data from a multitude of sources” via information brokers, to deliver more precise ads (Fourthwall Media, n.d.; Leahey, 2016; NCC Media, n.d.). NCC Media, the US cable TV ad platform owned by Comcast, Cox, and Spectrum, provided campaigns the ability to target potential voters via the integration of its set-top box viewing information with voter and other data from Experian and others (Miller, 2017). Deals between TV data viewing companies and organisations representing both Republican- and Democratic-leaning groups brought the “targeting capabilities of online advertising to TV ad buys…bringing what was once accessible only to large state-wide or national campaigns to smaller, down-ballot candidates”, explained Advertising Age (Delgado, 2016).

Psychographic, neuromarketing, and emotion-based targeting

Psychographics, mood measurement, and emotional testing have been used by advertisers for many decades, and have also been a core strategy in political campaign advertising (Key, 1974; Packard, 2007; Schiller, 1975). The digital advertising industry has developed these tools even further, taking advantage of advances in neuroscience, cognitive computing, data analytics, behavioural tracking, and other recent developments (Crupi, 2015). Granular-based messages that trigger a range of emotional and subconscious responses, to better “engage” with individuals and deepen relationships with commercial brands, have become part of the DNA of digital advertising (McEleny, 2016). Facebook (2015), Nielsen, and most leading brands use “neuromarketing” services worldwide, which utilise neuroscience tools to determine the emotional impact of advertising messages. There is a growing field, recently promoted by Google, of “Emotion Analytics” that takes advantage of “new types of data and new tracking methods” to help advertisers “understand the impact of campaigns—and their individual assets—on an emotional level…” (Kelshaw, 2017). Scholars have identified that the use of “psychological targeting” in advertising enables the influencing of large groups of people by “tailoring persuasive appeals to the psychological needs” of the specific audiences (Matz, et al, 2017). Data company Experian Marketing Services for political campaigns offered data that weaved together “demographic, psychographic and attitudinal attributes” to target voters digitally. Experian claims its data enables campaigns to examine a target’s “heart and mind” via attributes related to their “political persona” as well as “attitudes, expectations, behaviours, lifestyles, purchase habits and media preferences (Experian, 2011, 2015). One of the most well publicised and controversial players in the 2016 election was Cambridge Analytica (CA), a prominent data analytics and behavioural communications firm that claimed to be a key component in Donald Trump’s victorious campaign. The company used a “five-factor personality model” aimed at determining “the personality of every single adult in the United States of America” (Albright, 2016; Kranish, 2016).Known as OCEAN, the model rated individuals based on five key traits: openness, conscientiousness, extroversion, agreeableness, and neuroticism. Drawing from digital data, voter history, and marketing resources supplied by leading companies, including Acxiom, Experian, Nielsen, GOP firm Data Trust, Aristotle, L2, Infogroup, and Facebook, CA was able to develop an “internal database with thousands of data points per person”. Its research also identified key segments that were considered “persuadable”, and shaped the advertising content placed “across multiple digital channels (with the most effective ads also appearing on television) (Advertising Research Foundation, 2017; Nix, 2016). The strategy was based on developing messages that were tailored to the vulnerabilities of individual voters (Nix, 2016; Schwartz, 2017). CA has become the subject of much scrutiny and debate, and itself has made conflicting claims, with critics raising concerns over its techniques and expressing scepticism about the extent of its impact (Confessore & Hakim, 2017; Karpf, 2017). However, the company’s work was sufficiently convincing to the leading advertising industry research organisation, the Advertising Research Foundation (2017, March), that it honoured the firm with a “Gold” award in 2017 under its “Big Data” category.

Discussion

The above description provides only a brief overview of the data-driven marketing system that is already widely in use by candidate and issue campaigns in the United States. The increasingly central role of commercial digital marketing in contemporary political campaigns is reshaping modern-day politics in fundamental ways, altering relationships among candidates, parties, voters, and the media. We acknowledge that digital technologies have made important positive contributions to the vibrancy of the political sphere, including greatly expanding sources of news and information, significantly increasing opportunities for citizen participation, and empowering people from diverse backgrounds to form coalitions and influence policy. The same tools developed for digital marketing have also helped political campaigns substantially improve voter engagement, enhance their capacities for “small-donor” fundraising, and more efficiently generate turnout (Moonshadow Mobile, n.d.; Owen, 2017). However, many of the techniques we address in this paper raise serious concerns—over privacy, discrimination, manipulation, and lack of transparency.

Several recent controversies over the 2016 election have triggered greater public scrutiny over some of the practices that have become standard operating procedure in the digital media and marketing ecosystem. For example, “fake news” has a direct relationship to programmatic advertising, the automated system of “intelligent” buying and selling of individuals and groups (Weissbrot, 2016). These impersonal algorithmic machines are focused primarily on finding and targeting individual consumers wherever they are, often with little regard for the content where the ads may appear (Maheshwari & Isaac, 2016). As a consequence, in the middle of the 2016 election, many companies found themselves with ads placed on “sites featuring pornography, pirated content, fake news, videos supporting terrorists, or outlets whose traffic is artificially generated by computer programs”, noted the Wall Street Journal (Nicas, 2016; Vranica, 2017). As a major US publisher explained in the trade publication Advertising Age,

Programmatic’s golden promise was allowing advertisers to efficiently buy targeted, quality, ad placements at the best price, and publishers to sell available space to the highest bidders…. What was supposed to be a tech-driven quality guarantee became, in some instances, a “race to the bottom” to make as much money as possible across a complex daisy chain of partners. With billions of impressions bought and sold every month, it is impossible to keep track of where ads appear, so “fake news” sites proliferated. Shady publishers can put up new sites every day, so even if an exchange or bidding platform identifies one site as suspect, another can spring up (Clark, 2017).

Criticism from news organisations and civil society groups, along with a major backlash by leading global advertisers, led to several initiatives to place safeguards on these practices (McDermott, 2017; Minsker, 2017). For example, in an effort to ensure “brand safety”, leading global advertisers and trade associations demanded changes in how Google, Facebook and others conduct their data and advertising technology operations. As a consequence, new measures have been introduced to enable companies to more closely monitor and control where their ads are placed (Association of National Advertisers, 2017; Benes, 2017; IPA, 2017; Johnson, 2017; Liyakasa, 2017; Marshall, 2017; Timmers, 2015).

The Trump campaign relied heavily on Facebook’s digital marketing system to identify specific voters who were not supporters of Trump in the first place, and to target them with psychographic messaging designed to discourage them from voting (Green & Issenberg, 2016). Campaign operatives openly referred to such efforts as “voter suppression” aimed at three targeted groups: “idealistic white liberals, young women and African Americans”. The operations used standard Facebook advertising tools, including “custom audiences” and so-called “dark posts”—“nonpublic paid posts shown only to the Facebook users that Trump chose” with personalised negative messages (Green & Issenberg, 2016). Such tactics also took advantage of commonplace digital practices that target individual consumers based on factors such as race, ethnicity, and socio-economic status (Google, 2017; Martinez, 2016; Nielsen, 2016). Civil rights groups have had some success in getting companies to change their practices. However, for the most part, the digital marketing industry has not been held sufficiently accountable for its use of race and ethnicity in data-marketing products, and there is a need for much broader, industry-wide policies.

Conclusion

Contemporary digital marketing practices have raised serious issues about consumer privacy over the years (Schwartz & Solove, 2011; Solove & Hartzog, 2014). When applied to the political arena, where political information about individuals is only one of thousands of highly sensitive data points collected and analysed by the modern machinery of data analytics and targeting, the risks are even greater. Yet, in the United States, very little has been done in terms of public policy to provide any significant protections. In contrast to the European Union, where privacy is encoded in law as a fundamental right, privacy regulation in the US is much weaker (Bennett, 1997; Solove & Hartzog, 2014; U.S. Senate Committee on Commerce, Science, and Transportation, 2013). The US is one of the only developed countries without a general privacy law. As a consequence, except in specific areas, such as children’s privacy, consumers in the US enjoy no significant data protection in the commercial marketplace. In the political arena, there is even less protection for US citizens. As legal scholar Ira S. Rubinstein (2014) explains, “the collection, use and transfer of voter data face almost no regulation”. The First Amendment plays a crucial role in this regard, allowing the use of political data as a protected form of speech (Persily, 2016).

The political fallout over the how Russian operatives used Facebook, Twitter, and other sites in the 2016 presidential campaign has triggered unprecedented focus on the data and marketing operations of these and other powerful digital media companies. Lawmakers, civil society, and many in the press are calling for new laws and regulations to ensure transparency and accountability for online political ads (“McCain, Klobuchar & Warner Introduce Legislation”, 2017). The U.S. Federal Election Commission, which regulates political advertising, has asked for public comments on whether it should develop new disclosure rules for online ads (Glaser, 2017). In an effort to head-off regulation, both Facebook and Twitter have announced their own internal policy initiatives designed to provide the public with more information, including what organisations or individuals paid for political ads and who the intended targets were. These companies have also promised to establish archives for political advertising, which would be accessible to the public (Falck, 2017; Goldman, 2017; Koltun, 2017). The US online advertising industry trade association is urging Congress not to legislate in this area, but to allow the industry to develop new self-regulatory regimes in order to police itself (IAB, 2017). However, relying on self-regulation is not likely to address the problems raised by these practices and may, in fact, compound them. Industry self-regulatory guidelines are typically written in ways that do not challenge many of the prevailing (and problematic) business practices employed by their own members. Nor do they provide meaningful or effective accountability mechanisms (Center for Digital Democracy, 2013; Gellman & Dixon, 2011; Hoofnagle, 2005). It remains to be seen what the outcome of the current policy debate over digital politics will be, and whether any meaningful safeguards emerge from it.

While any regulation of political speech must meet the legal challenges posed by the First Amendment, limiting how the mining of commercial data can be used in the first place can serve as a critically important new electoral safeguard. Advocacy groups should call for consumer privacy legislation in the US that would place limits on what data can be gathered by the commercial online advertising industry, and how that information can be used. Americans currently have no way to decide for themselves (such as via an opt-in) whether data collected on their finances, health, geo-location, as well as race or ethnicity can be used for digital ad profiling. Certain online advertising practices, such as the use of psychographics and lookalike modelling, also call for rules to ensure they are used fairly.

Without effective interventions, the campaign strategies and practices we have documented in this paper will become increasingly sophisticated in coming elections, most likely with little oversight, transparency, or public accountability. The digital media and marketing industry will continue its research and development efforts, with an intense focus on harnessing the capabilities of new technologies, such as artificial intelligence, virtual reality, and cognitive computing, for advertising purposes. Advertising agencies are already applying some of these advances to the political field (Facebook, 2016; Google, n.d.-a; Havas Cognitive, n.d.). Academic scholars and civil society organisations will need to keep a close watch on all these developments, in order to understand fully how these digital practices operate as a system, and how they are influencing the political process. Only through effective public policies and enforceable best practices can we ensure that digital technology enhances democratic institutions, without undermining their fundamental goals.

This paper is part of 'A Manchurian candidate or just a dark horse? Towards the next generation of political micro-targeting research’, a Special issue of the Internet Policy Review.

Introduction

As political campaigns compete, they try to outsmart each other by all sorts of actions: from dropping witty puns during a televised debate, to strategically knocking on doors and convincing voters. Technological innovation can help political parties improve the effectiveness of their campaigns. By using technology to collect, process, and analyse information about voters, campaigns can improve their knowledge about the electorate. Subsequently, technology can extend campaigns' capabilities of targeting specific groups with tailored messages resulting in more efficient campaigning. We call this phenomenon 'political behavioural targeting' (PBT).

Several scholars have researched political behavioural targeting in the US context (e.g., Kreiss, 2012, 2016; Nielsen, 2012; Hersh, 2015). However, the US differs in several obvious ways from most European countries. One can imagine that differences in electoral systems, privacy laws, and party financing influence campaigns’ ability to collect, process, and use personal voter data. Therefore, the findings from these studies do not necessarily apply to European countries. As there is little research in a European context, it remains unclear to what extent and how campaigns in a multiparty democracy, such as the Netherlands, use PBT-techniques. Also, it is unclear if and why there are differences between parties. In line with Colin Bennett (2016, p. 261), we wonder: "can political parties campaign in Europe as they do in North America?"

Such a question is relevant, as some scholars fear that the use of data and targeting techniques hinders public deliberation (Gorton, 2016), weakens the mandate of elected officials (Barocas, 2012), has negative effects on citizens’ privacy (Howard, 2006; Rubinstein, 2014; Tene, 2011), and enables campaigns to send tailored messages directly to citizens, thereby avoiding scrutiny from journalists (Jamieson, 2013). As a result, campaigns can potentially make opposite promises to different people, without anyone noticing.

This article sheds light on how Dutch political campaigns adopt and use PBT-techniques. Through interviews with campaign leaders, using a grounded theory approach, we answer the following overarching research question: What barriers and facilitators for the adoption and use of PBT-techniques do Dutch political parties perceive?

Theoretical framework

We will first summarise innovations in political campaigns over time, leading up to the advent of political behavioural targeting. Then, we identify the factors influencing the adoption of PBT on a campaign team level. Finally, we explore the factors that can shape the adoption of PBT on the level of national systems.

Innovations in political campaigns

Political campaigns have continuously been adapting to technological developments. Pippa Norris (2000) describes how the advent of television and the shift from partisan newspapers to national television news triggered a process of modernisation in the way political campaigns operated. Notable consequences of this shift were the adoption of a media-centred strategy in order to set the agenda, the rise of political marketing, the collection and use of data (such as opinion polls) to "shape, fine-tune and monitor campaign efforts" (Blumler, Kavanagh, and Nossiter, 1996; p. 53). Another shift came with the internet and the new possibilities for party-voter interaction that came along with the medium, which led campaigns to a new stage of the modernisation process: the postmodern campaign (Norris, 2000).

It would be an oversimplification to point to 'the internet' as a game-changer in political communication, because of the rapidly changing nature of the internet itself. As David Karpf (2012, p. 640) notes: "the internet of 2002 has important differences from the internet of 2005, or 2009, or 2012". Accordingly, much more than the advent of the internet itself, it is the advent of social media such as Facebook (2004), YouTube (2005) and Twitter (2006) which provided political campaigns with new ways of communication with the electorate (e.g., Gibson & McAllister, 2011; Conway, Kenski, & Wang, 2015; Vaccari, 2012). Together with companies such as Google, whose core business is actually not its well-known search engine but rather its advertisement business, social media not only facilitate new ways of communication, but also the tracking and collection of behavioural data of internet users (Zuiderveen Borgesius, 2016). This technique ("behavioural targeting") originates from the advertisement business. Ad agencies monitor people's online behaviour and combine this information with consumer data provided by data brokers, to target them individually with tailored ads (Turow, 2011, p.75). When applying this concept to the political realm, we can dub this phenomenon as political behavioural targeting (PBT).

Of course, PBT is not about selling products but about winning votes. And political campaigns have different means to do so than advertisement agencies have (e.g. canvassing efforts); which means that PBT happens offline as well as online. We distinguish traditional canvassing from PBT-canvassing if campaigns are able to process information about individual conversations (such as the voter's likelihood to vote for a party or her most important voting consideration), and subsequently use that information to gain strategic insights about the electorate and/or to target the voter at a later stage with a tailored message, while skipping the 'wrong' doors in a neighbourhood (Kreiss, 2016; Nielsen, 2012).

Arguably, the use of PBT can be seen as the latest step within the modernisation of political campaigns. However, as we have seen in earlier phases, not all parties in all countries adopt new techniques at the same pace and rate. Below, we identify the factors influencing the adoption of PBT. We organise these factors at two levels: (1) the individual campaign around a candidate/party and (2) the national system (i.e., the electoral system, regulatory framework, and culture). This translates into the model shown in Figure 1, which will be elaborated on in the next paragraphs.

Figure 1: Factors influencing the adoption of PBT

The campaign team level

In his extensive research of US political campaigns, Daniel Kreiss (2016) identified four factors concerned with technological innovation within political campaigns. There are resource factors, such as campaign budgets and the number of volunteers a campaign can employ; infrastructural factors, such as technological tools or skills within the organization; organizational factors, such as organisational culture and structure; structural electoral cycle factors, such as election results. Building upon Kreiss' factors, we add an additional four (one campaign team level factor and three system level factors) to examine the use of PBT. On a campaign team level, the factor is ethical and legal concerns, such as normative reservations towards PBT. On a system level, the factors are electoral context, regulatory framework,and culture (discussed below). These new factors were identified through a review of literature about innovation in data-driven political campaigning techniques (e.g., Anstead, 2017; Kreiss, 2016; Jungherr, 2016; Hersh, 2015; Nielsen, 2012), and literature about (hybridisation of) campaign evolvement (e.g., Lijphart, 2012; Plasser & Plasser, 2002; Karlsen, 2010; Norris, 2000).

Resource factors

The main elements within this factor that could influence the extent to which campaigns can use PBT-techniques are: the budget and the effort needed to carry out a PBT-operation. A large budget enables campaigns to hire skilled personnel, acquire data, or buy targeted ads. The same dynamic applies to the number of volunteers a campaign can mobilise: having a lot of them facilitates a campaign in collecting data by canvassing, and sending potential voters targeted messages (the use of volunteers, of course, is dependent on their skills). Having a small budget and few volunteers, consequently, can be a barrier for campaigns because it bars them from acquiring the same amount of capabilities or from carrying out an operation on a large scale. This is in line with normalization theory (Margolis & Resnick, 2000), according to which the possibilities of the internet will not upset traditional power structures, but will rather develop along traditional lines as in the 'offline world'.

We can also view PBT as a means of using a campaign's resources as efficient as possible, to ensure parties do not spend money and effort on voters who will vote for another party anyway, or on citizens who will not vote altogether. Then, parties with limited resources could be more inclined to use PBT to not waste precious money, time, and labour. This is in accordance with the idea of equalization, which views the internet as an empowering tool for smaller parties due to its low costs and its new ways of direct communication with the electorate (Margolis, Resnick, & Levy, 2003; Bimber & Davis, 2003; Stanyer, 2010). A meta-analysis found evidence for the existence of both normalisation and equalisation in election campaigns (Strandberg, 2008). The occurrence of either process can differ per country and is dependent on several contextual factors, which will be discussed later on.

Organisational factors

The elements in this factor are about how campaign leaders perceive campaigning. Do they rely on proven best practices from previous campaigns or is there a culture of innovation? John Padgett and Walter Powell (2012) describe the concept of network folding. Applied to the political realm, this entails the extent to which campaigns employ skilled personnel from non-political sectors and to integrate that expertise into their existing institutions. An example is the hiring of Google engineer Stephanie Hannon as chief technology officer by the Clinton campaign (Easton, 2015). The 'cognitive diversity' following from network folding can lead to creative ideas (De Vaan, Stark, & Vedres, 2015). Furthermore, the organisational structure can be expected to resemble the way the campaign perceives PBT. A campaign with an autonomous data department is probably more prone to rolling out a PBT-strategy than a campaign that sees 'data' as only one of the many tasks of a communication staffer. Also, a change in leadership can be a facilitator for innovation (Gibson & Römmele, 2001).

Infrastructural factors

Elements are the technological tools available to campaigns, which enable them to roll out a PBT-operation. For instance, such tools might assist volunteers in the field by enabling them to collect data. They can be developed in-house or outsourced; in fact, there are specialised third party consultancies, that offer off-the-shelf tools, which in turn allow campaigns to employ innovative technology even though the campaigners do not have any technical expertise.

Structural electoral factors1

The actions of rival campaigns fall under the umbrella of structural electoral factors. A successful PBT-campaign of a rival can facilitate innovation in other campaigns, especially if those other campaigns themselves look back at an unsuccessful election. This connects with the 'critical event' (Kreiss, 2016), such as losing an election that should have been won, or with the experience of an 'external shock', which can be an incentive for professionalisation (Gibson & Römmele, 2001).

A second element influencing campaigns' likelihood to use PBT-techniques, is issue ownership (Petrocik, 1996), and the subsequent statements of party candidates propagating standpoints of the party. A political campaign 'caught' using privacy-infringing PBT-techniques, while its candidates present themselves as privacy champions, is likely to come across as hypocritical. Being perceived as such should be avoided, considering the negative electoral consequences following political-ideological hypocrisy perpetrated by politicians (Bhatti, Hansen, & Olsen, 2013).

Ethical and legal concerns

Elements within this factor consist of ethical and legal restrictions on how campaigns operate. For example, a political party could believe that PBT is ethically wrong as it infringes on citizens' right to privacy, and citizens' autonomy to form their own opinions. As a result, the party 'self-regulates' and refrains from using campaigning techniques violating its ethical beliefs.

Another element is the legal uncertainty that occurs when a campaign does not know how to behave in accordance with data protection and election laws, because of a lack of internal expertise. Such confusion can result in differences in the actions taken by comparable actors (e.g. Raskolnikov, 2017). Legal uncertainty can lead to 'overcompliance', which can be seen as a barrier towards the adoption of PBT-techniques, or to 'undercompliance', which facilitates the adoption of PBT-techniques (Calfee & Craswell, 1984). For instance, Anstead (2017) notes how parties felt disadvantaged by targeting possibilities facilitated by the perceived undercompliance with UK campaign finance law during the 2015 general elections.

The system level

Aside from campaign level factors, we look at contextual factors as well. These factors may limit the extent to which (US-American) campaigning techniques can be adopted in other countries (Karlsen, 2010). Therefore, we add three new contextual factors to our model. We expect that the electoral system, the regulatory system, and the culture of a democracy influence the extent to which the campaign team level factors are applicable. Below, we explore how the adoption of PBT-techniques can be influenced by properties of different systems. We will later apply our model (see Figure 1) to one specific case.

Electoral system

The three dominant electoral systems are first-past-the-post (FPTP), proportional representation (PR), and two-round (TR) (Birch, 2001, 2003). How these systems function, can influence how campaigns are run. The FPTP-system, first, can lead to an overvaluation of some key districts. Such districts sometimes 'swing' to one party and sometimes to another party, whereas other districts go to the same party in each election. As an effect, campaigns in a FPTP-system are inclined to spend a disproportionate amount of money and labour in these key districts in the hope of swinging the election their way (e.g., Anstead, 2017; Lipsitz, 2004). The PR-system, second, does not favour a select group of voters in a few key districts (Plasser & Plasser, 2002). This is especially true when the PR-system consists of only one district, in which every vote counts equally. As a result, campaigns have to spread their means more equally over the country. The TR-system, third, makes for a relatively unpredictable campaign, since it often is unclear which candidates will make it to the second round. Furthermore, the TR-system makes it important for campaigns to collect the votes of the supporters of the losing candidates of the first round. Therefore, campaigns should not only focus on their own base but other candidates' bases as well (Blais & Indridason, 2003). This has consequences for PBT, since campaigns should not only correctly classify potential voters as their own, but the other voters as well in order to target them in the next round.

A different aspect of electoral systems that influences how a campaign is run, is the degree of fractionalisation in a democracy (Duverger, 1959; Lijphart, 2012; Wang, 2012). FPTP-systems favour relatively few candidates/parties. PR-systems, in contrast, enable a large number of parties to run in an election. The first round of a TR-system can consist of many different candidates. As a result, campaigns that operate in a PR or a TR-system are less likely to launch attack campaigns against competitors. This is because PR-systems generally require a coalition of parties working together after the elections (Plasser & Plasser, 2002). And in a TR-system, campaigns should not mistreat competing candidates too much because winning campaigns have to court the bases of losing candidates in order to win the second round. Furthermore, in a highly fractionalised democracy, parties represent different (minority) groups within the electorate. This results in a high risk of 'mistargeting', in which campaigns approach a member of group A with appeals for a member of group B. Mistargeting can lead to voters penalising the campaign for their mistake (Hersh, 2013). These contextual circumstances may call for different PBT-strategies.

Regulatory framework

We distinguish strictly regulated campaign environments, moderately regulated campaign environments and minimally regulated campaign environments (Plasser & Plasser, 2002). Strictly regulated environments are characterised by "severe restrictions on the contact and communication between candidates and their constituencies" (p. 137). Moderately regulated environments typically focus on regulating access to TV-advertising and campaign funding. Minimally regulated environments impose few regulatory restrictions on political campaigns. It may be infeasible to implement innovative PBT-techniques in strictly regulated environments. Legal uncertainty can play a role on a system level too (e.g. because of a gap in the law).

Culture

Differences in the adoption and use of innovative PBT-practices can also be influenced by the culture or tradition in a democracy. For example, turnout culture is important because campaigns operating in countries where turnout is high will focus more on convincing voters on getting out the vote than campaigns operating in a low-turnout culture. In a low-turnout culture, campaigns sometimes target specific groups of voters (e.g. the elderly, who are more likely to turn out) more than other groups (e.g. the poor), who are unlikely to turn out (Herrnson, 2001). The turnout culture can influence the data campaigns collect on someone (and how campaigns tailor their messages), because a campaign message meant to convince someone typically leans on more data than a message meant to mobilise a voter does. Furthermore, cultural norms can dictate the strategy of political campaigns. In Japan, for example, posting dark post attack ads, such as the 'super predator' ad Trump launched against Clinton (Green & Issenberg, 2016), is improbable because of the cultural convention of averting direct conflict (Plasser & Plasser, 2002).

System level context is likely to affect campaign level factors. A campaign operating in a multiparty PR-system needs to pour more resources into identifying potential supporters than a campaign in a FPTP-system. After all, identifying potential Republicans or Democrats is easier than identifying potential voters in a ten-party race. Moreover, other than in US campaigns, most European campaigns are unable to access voter registration files provided by an electoral register. In many countries citizens can just show up at the voting booth, which means that the whole act of ‘registering’ for voting, as it is the case in the US, does not exist. Since Hersh (2015) has found that voter lists are one of the most valuable pieces of data to US campaigns, this principal unavailability, or non-existence, of such data poses a challenge for the PBT-capabilities of campaigns. But this challenge should by no means imply a PBT-operation in Europe is impossible. We would argue that while the lack of access to voter lists makes it very difficult to achieve the same level of granularity when engaging in PBT as in the US, by using other commercially available or self-collected data, it can be possible to come reasonably close to the desired level (which may be more modest for European campaign leaders than for American ones). The extent to which there is an actual difference between the degree of granularity between US and European campaigns, however, is outside of the scope of this paper, as we focus on perceptions and strategies of campaign leaders.

System level context also affects infrastructure. For instance, should the groundwork be spread equally across the nation, or focused on a number of battleground states? Also, a campaign operating in a heavily regulated context is likely to encounter some legal barriers. For example, because of campaign financing regulations (may influence resource factors), and data protection regulations (infrastructural factors). The absence of regulations, conversely, can facilitate PBT. Cultural context, finally, can influence campaigns' ethical considerations regarding PBT. Campaigns operating in a culture that favours privacy, for example, can be expected to avoid (or use less-invasive) PBT-techniques than campaigns run in a culture in which privacy is less important. In sum, there are several factors, both on campaign and system level, which can form a barrier or facilitate the extent to which campaigns are able to use PBT-techniques and how they use them.

Extending existing research to a European context, we have developed and will apply an improved model (applicable in different electoral contexts) to analyse barriers and facilitators to innovative PBT-practices by political campaigns. As the context of the research case differs from the US, we expect to contribute to the framework and to shed light on how contextual factors influence innovation of political campaigns. Furthermore, in answering our research question, we provide insight into the way political campaigns in a multiparty democracy organise, communicate and innovate. Given these considerations, our key question is: What barriers and facilitators for the adoption and use of PBT-techniques do Dutch political parties perceive?

Method

This study focuses on campaigns in the Netherlands because of the national elections taking place in the research period (15 March 2017), the advanced technological infrastructure (Coy, 2015), and the interesting contextual factors. The Dutch electoral system is one of open list proportional representation (PR), in which all members of parliament come from one nationwide district (Lijphart, 2012). This means that in the Netherlands, every vote counts equally. Moreover, the system of PR (and the very low de-facto threshold) enables a relatively large number of political parties to run in an election. 28 parties participated in the 2017 national election ("Partijen nemen deel", 2017). Of these parties, 13 actually gained a seat in parliament ("Officiële uitslag", 2017).

The Dutch national elections have a relatively high turnout: around 80% in the previous two elections ("Officiële uitslag", 2017). But where the US presidential campaigns can spend hundreds of millions of dollars (Narayanswamy, Cameron, & Gold, 2017), the Dutch campaign with the biggest funds (VVD) has no more than 5 million dollars to spend. And even if the budgets were sufficiently large, the question is whether voter data would be usable for a political campaign. The Dutch data protection law categorises political preference as sensitive personal data. This means that campaigns are only allowed to process such information if the potential voter explicitly gives permission to do so.

Finally, as party membership in the Netherlands steadily decreases, political campaigns can rely less on their members to do labour-intensive tasks (such as canvassing). In 2016, the number of party members of all political parties combined, was at its lowest point since the second world war. Although this number has picked up slightly since, party membership is still quite low ("Membership Dutch parties still low", 2017).

After approval from the ethical committee of the University of Amsterdam, we carried out eight in-depth interviews with campaign leaders. We interviewed 11 campaign leaders, belonging to eight political parties in total (three interviews were double-interviews). In addition to this, we held two background interviews (with one local campaigner for the municipality of Amsterdam and one political consultant offering PBT-services). The eight elite-interviews on average lasted 53 minutes. Two were conducted by phone, the others face-to-face. We took a qualitative research approach for several reasons: the small group of people concerned with the coordination of political campaigns in the Netherlands, the lack of knowledge on this topic in the Netherlands, and because interviewing is a suitable method for understanding the mechanisms behind and perceptions of a phenomenon (Boeije, 2005). As we want to understand how campaigns see PBT, what they are actually doing, and how they perceive possible barriers and facilitators to the adoption and use of PBT-techniques, the interview is a suitable data-collection method. Using an interview guide (see appendix A), we held semi-structured interviews, allowing follow-up questions.

Interviewees

We selected the interviewees via purposive sampling. Campaign leaders qualified for an interview when they had a coordinating role in the campaign and were campaigning for a party that gained at least one seat in the 2012 national parliamentary elections. Eleven campaigns satisfied this second criterion (see Table 1). We contacted interviewees via email, explaining the objective of the study. The interviewees signed an informed consent document before the interview started. We also promised the campaign leaders anonymity, and confidentiality until after election day (15 March 2017). By doing so, we tried to provide the interviewees with a safe environment in which they felt like speaking freely, and without concern of somehow 'leaking' strategic information. Because no information would become public before election day, the risk that interviewees might provide biased information due to a strategic agenda was minimised. Another large advantage of interviewing the campaign leaders before the elections took place, is the prevention of hindsight bias by the interviewees. Unfortunately, we were unable to convince three parties to comply (VVD [right on the political spectrum], PVV [right-wing nationalist party], PvdD [Party for the animals; left-wing]). These three parties were unwilling to cooperate, either because they still found the risk of leaking their strategy too large, or they did not offer an explanation.

Analysis

Using a grounded theory approach, this study has passed four phases: the exploration phase, the specification phase, the reduction phase, and the integration phase (Wester, 1995). In the exploration phase, two background interviews took place (with a campaigner for the municipality of Amsterdam and with a political consultant offering PBT-services). These were coded using AtlasTI, 'tentatively labelling' relevant information (Glaser, 1978). Thereafter, the first interviews with campaign leaders took place. These were transcribed and open-coded. Furthermore, fellow researchers also coded these interviews and discussed the content (peer debriefing). In the next phase, new interviews took place and the data was subject to axial coding. The first dimensions were identified (e.g., what forms a barrier and what facilitates the use of PBT-techniques?). The reduction phase saw the emergence of the core category (innovation). In the integration phase, we completed the conceptual framework, finalised our analysis, and had the campaign leaders approve the quotes used (member checking). This means the campaign leaders agreed with the way they were quoted, and with the publishing of the names of the political parties. Member checking increased the willingness of campaign leaders to cooperate with the study.

Results

We first describe the field: to what extent do campaigns use PBT-techniques? Then we explain differences between parties by focusing on the five campaign level factors concerning the use of PBT (resource, infrastructural, organisational, structural electoral cycle factors, and ethical and legal concerns). Finally, we zoom out to the system level and discuss the influence of contextual factors (electoral system, regulatory framework and culture) on the adoption of PBT-techniques.

PBT in Dutch campaigns

As campaigns in the Netherlands can have recourse to relatively detailed public census data and detailed election results, all campaigns, to some extent, adopt a PBT-approach. Furthermore, Facebook is an important tool for all parties, but the parties differ in how they use Facebook's capabilities. Some parties occasionally post content targeted to broad age groups, while other campaigns frequently post content tailored to more specific groups. Two campaigns stand out, as they have developed their own PBT-tools, which they can use to continuously refine their knowledge of the electorate. We will now use our model to explain the differences between campaigns.

Resource factors

All campaigns cite financial costs as a barrier. Table 2 shows that budgets are modest, and differ between parties.

Source: parties' annual financial reports, on file with authors.

These small budgets form a barrier for the cooperation with expert political consultants (such as Blue State Digital) to enhance their PBT-operations. Parties refer to the financial costs as the main reason not to hire consultants. Campaign leader 1 of the Social Democratic Party (PvdA) explains why he does not work with Blue State Digital (BSD):

Their system is very expensive, that's a factor. And you need the people to carry out the work for you. In an ideal world, such a cooperation would be really cool though."

Liberal democrat party D66 agrees: "because it costs a lot of money and we don't have that kind of money. And if we spend it on a consultant, we can't spend it on the campaign itself."

The same barrier appears when campaigns speak about other technological means, such as canvassing apps, allowing campaigns to directly process information from canvassers. Christian Democrats CDA, for instance, would like such an app. Campaign leader 8: "yes, but that would demand a financial investment that we can't afford."Green Party GroenLinks has a contrasting perspective: "I believe it usually costs around €100,000 to build an app such as our own. (..) We, however, paid our programmers two crates of beer and 40 pizzas." Several facilitators help GroenLinks and also socialist party SP to overcome this barrier of financial costs. First: the personal network of the campaign leader. This facilitator is especially prominent for GroenLinks, where campaign leader 7 employs his own network to optimise the BSD-systems, but also to help him with setting up other parts of the campaign:

We had to adjust it [the BSD system] somewhat for the Netherlands. The people with whom I did so, Swedish folks ... they are simply a little network of people of around my own age, and some people who are a bit older and have already set up a similar campaign in their own countries. A guy who set up the grassroots organization for Trudeau, for example, he's a couple of years older than I am, but I Skype with him to talk about how I should handle certain things.

Campaign leader 7's personal network plays (or at least played)an important role in cheaply setting up technological tools and creating content:

Through the network, I'm aware of the crowdfunding streams for a normal campaign. (..) I'm meeting a friend tomorrow, who has experience with mail flows. (..) I have a network of volunteering writers, poets, freelance journalists who write for us for free. (..) So partly, I just have a good personal network.

However, as PvdA notes, having lots of data is of no use if you don't have the capacity to use it. GroenLinks tries to overcome this barrier by organising their campaigns, to some extent, in a citizen-initiated manner (Gibson, 2015). A citizen-initiated campaign (CIC), devolves "power over core tasks to the grassroots" (p.183). As campaign leader 7 puts it:

Grassroots is about creating an infrastructure to enable as many sympathizers as possible to volunteer as canvassers on a large scale. So voter contact on a large scale, but also – and that's Bernie's [Sanders] lesson – to have places in which a few people make stuff by themselves without us having any control over it. (..) Embracing people's creativity without managing it.

SP has less need of a citizen initiated campaign, because of their relatively large number of active party members. "The big difference [with GroenLinks] is the fact that we already have the volunteers. Many other parties lack the numbers. We have thousands of party members who gladly canvass for us two weekdays and on Saturday as well." [Campaign leader 11, SP].

Infrastructural factors

Having a good infrastructure allows campaigns to actually collect data and send tailored messages. What kind of PBT-infrastructure can parties rely on and how does it facilitate their use of PBT?

All campaigns use the PBT-infrastructure Facebook offers, although some more than others. Nearly all campaigns use its lookalike audiences function to find new potential voters. Campaign leader 8: "we search for profiles of people who look like the ones who’ve already liked our Facebook page, and then serve them with advertisements." Campaigns also look at people who like pages that are close to the values of the political parties. Christian party ChristenUnie, for instance, tries to target voters who like the page of evangelical broadcaster EO. So does the Calvinist Political Party SGP, which tries to find out people’s interests on Facebook: "For example… farming, or Israel, off the top of my head; you try to approach people along the lines of their interest, or the region in which they reside." [Campaign leader 9]

Some campaigns also employ ‘dark posts’, a Facebook function that enables campaigns to opaquely target specific audiences, while its messages are not visible to untargeted Facebook users. Campaign leader 1 exemplifies:

We’ve managed to get something done related to gas extraction in Groningen. It doesn’t make sense to share that on the national Facebook page, because it was only important news locally. So we put out a dark post, only for Groningen residents. Sometimes we can specify it even more.”

Using Facebook for PBT-purposes, campaigns do not actually gather or own data themselves. There are a few campaigns that do gather their own data, by using canvassing apps. Campaign leader 7:

We use the election results per voting location and use that information to establish the GroenLinks mindedness of a neighbourhood. Then we can prioritise which addresses to visit and which to ignore. When we visit addresses, our volunteers use the app to answer the following questions: 1. Is anyone home? 2. Does she want to talk? 3. Is she going to vote? 4. Is she planning to vote for GroenLinks? 5. What is the most important theme to her? 6. How GroenLinks minded was she? If she considers to vote for GroenLinks, two questions follow: 1. Do you want to stay informed of our campaign by e-mail? 2. Can I have your phone number, so we can ask you to do canvassing talks?”

The GroenLinks app facilitates large scale collection of information about people's political preferences, thereby informing strategic decisions. Also, the personal data can facilitate accurate PBT on an individual level. The secondary objective of the app is to provide an infrastructure for volunteers to campaign on their own terms, whenever they feel like doing so:

Our app, built by hackers, enables others to campaign for us. (..) Someone in [small town] Lutjebroek can install our app and go ahead and work for our campaign. No campaign leader needed. [Campaign leader 7]

Some campaigns monitor the visitors of their own websites. Campaign leader 1: “What are people searching for on our website, how do they get to our website, how much time do they spend, (..) which button should you colour red? How does that work?” At the time of the interview, CDA was not yet tracking their website visitors, but: “we’ve just migrated to a new website, on which we want to start collecting more data on our visitors. I’m curious what kinds of people are visiting the website. And what kinds of people don’t, and therefore have to be reached through different channels.”

SP has built a system which combines previous election results, census data and their own membership Constituent Relationship Management (CRM) data. Plotted on a Google Map, they can identify interesting areas for them to canvass. This system facilitates efficient use of means:

We would do nothing more happily than knocking on every single door in every city, but unfortunately, we do not yet have that kind of manpower. So we do an analysis: What kinds of neighbourhoods are especially interesting for us? We have built our own system to help us make that decision [Campaign leader 11].

Organisational factors

Circumstances within the campaign’s organisation itself can form a barrier for the uptake of PBT-techniques. Less innovative parties, for instance, do not have a dedicated data, tech, or digital department. As a campaign leader notes: "The department responsible for that [tech/data/digital] is our Communication department. So that's four or five people. And sometimes someone of the department picks it up, but there's not one specific person who's responsible."This contrasts with GroenLinks, which has a Digital and Grassroots department and with SP's Digital department.

The "state of mind"within a campaign can also be seen as a barrier:"In the sense that internally, people are still very much inclined to think offline. The culture within the campaign is quite offline." [anonymous campaign leader]2

New leadership and younger staffers can play a facilitating role in political organisations. Campaign leader 11 argues that, because he is young, their new party chair brings a more tech-savvy vision than his predecessor. According to the campaign leader, younger staffers are more likely to implement tech and data in their work procedures.

A final organisational barrier is the primary goal a political party pursues. Campaign leader 10:

Maybe the strange thing about SGP is that we do not care that much about seat maximization. For us, it's about the impact of our principles. And sure, we would rather have four seats than three, but if we have to settle for three seats: that's fine too. And that's, in my opinion, a reason why we have a feeling like: do we really need data?

Structural electoral cycle factors

These circumstances are largely beyond the control of the campaigns, but they can influence the uptake of PBT-techniques. Campaign leaders see the PBT-actions of other political campaigns as a motivational factor. As campaign leader 11 notes about the development of their app: "I've looked a little bit at how GroenLinks have their app and canvassing system." Or as campaign leader 2 concludes: "If every party does it, you don't win very much by it. But if you're the only party that does nothing..."

Ethical and legal concerns

Especially D66 and the seniors’ party 50PLUS take a principled stance against the collection of data and the use of PBT. Where D66 presents itself as a privacy champion and therefore will never gather and use information about (groups of) voters, 50PLUS campaign leader 6 warns about the risk of irresponsible use of the data gathered by the "almost stalking of people", which he calls "morally irresponsible".

Furthermore, a lack of internal legal expertise appears to contribute to a feeling of legal uncertainty, which affects the likelihood of adopting PBT-techniques: "Legislation has grown so very comprehensive and complex. It's almost impossible to cope for us as a small organisation." [Campaign leader 5]

While ethical and legal concerns can form a barrier, a left- or right-wing orientation does not seem to be instrumental therein. After all, we have seen left-wing parties GroenLinks and SP develop relatively advanced PBT-tools. And we have seen right-of-centre party CDA express clear interest in advancing their own PBT capabilities. At the same time, left-of-centre 50PLUS and right-of-centre D66 both oppose the use of PBT.

System level

Electoral system

Although the Dutch one district PR-system should make for a rather equal distribution of campaign efforts, campaigns still divide the country into smaller areas of interest called 'key areas'. These areas differ per party, but do receive a relatively large part of campaign attention. Campaign leader 1 describes these as areas: "where we know the turnout is low, but the number of PvdA-voters is high". All campaigns use data provided by the Electoral Council, showing the election results per party, per voting location to establish key areas. Campaign leader 3 explains:

Using that [the election results], you see: Okay, we do well in this neighbourhood or this street. And then you combine that information with the CBS3 data, to find out what kind of neighbourhood it is, what kind of people live there, what are their backgrounds, how much do they earn, what does the family composition look like, et cetera.

Facilitated by these public data, campaigns enrich their knowledge of specific areas. A next step would be to use those data to make personalised appeals to (subgroups of) people living in those specific key areas.

Regulatory framework

Although the Netherlands would qualify as a minimally regulated environment (Plasser & Plasser, 2002; Esser & Strömbäck, 2012), campaigns all experience regulatory pressure and legal uncertainty on a system level. They cite an abundance of regulations, forming a barrier to their ability to innovate.

The technological developments have been taking place so very quickly. And, in that timeframe, to adjust all your procedures and everything. And also to meet the privacy regulations, I think many parties face a huge challenge in that respect." [Campaign leader 4]

Campaigns sometimes face a dilemma, having to decide between innovative techniques and privacy regulations. Campaign leader 11:

Regulations sometimes are unclear, which leads us to decide to go for the safe option because you do not know where the red line is. And you never want to abuse someone's personal data. So yes, regulations sometimes cause us to hit the brake and that's a good thing.

Culture

There is a recurring worry about the perceived low level of political knowledge of the average Dutch voter. PBT-techniques can facilitate campaigns' efforts to convince or educate such low-information voters, for example by "having a conversation with someone, especially if you share some characteristics," [Campaign leader 7] or by interesting "people for things that are relevant to them and to make them aware of the political dimension of those things." [Campaign leader 8] Campaign leader 3, in contrast, concludes that the electorate's low level of political knowledge (together with the perceived volatility of the electorate, the decrease in political trust, and their focus on persons instead of parties) forms an insurmountable barrier, making PBT-techniques irrelevant.

Discussion and conclusion

In the 2017 elections, used here as a case study, all campaigns use PBT through Facebook, but some parties are more advanced than others, and have even developed their own PBT-tools. We have established what the main barriers and facilitators for PBT are, using five factors on a campaign level and three factors on a system level. Not only does this study shed light on the conditions under which these barriers and facilitators manifest themselves, it also gives insight into their different workings across parties. Our study provides information about the data collected by parties and the PBT-techniques used to attract voters. We demonstrate how personal networks and cognitive diversity within a campaign can level barriers. We show how PBT is not only perceived as useful for campaigns in a FPTP-system, but in a PR-system as well. And we show how regulatory pressure is perceived as an obstacle and as a welcome 'normative red line'.

A triangulated research approach can improve our understanding of the campaign leaders' constructs. Observation of their (use of) PBT-tools and how these tools help campaigns make strategic decisions, can give more insight into the workings of these techniques. Another approach would be to interview canvassers and identify 'field-level' barriers and facilitators. Furthermore, ideally, we would have spoken to all parties holding a seat in parliament. Unfortunately, three parties did not cooperate. Two of those became the largest (VVD) and second largest (PVV; in a very close field) party. Since we did have access to eight of 11 parties, we are confident about our findings and we do not expect to identify additional factors influencing the adoption of PBT from interviews with the remaining parties.

Compared to related recent studies by Anstead (2017), Hersh (2015), Kreiss (2016), and Nielsen (2012), this study makes a number of contributions. In general, we focus our exploratory research on a PR-system instead of a FPTP-system, and we develop a model that takes system level contextual factors into account. Specifically, unlike Anstead (2017) we have found evidence for equalisation (which occurs when smaller parties take advantage of the internet's low costs and direct communication possibilities, and, in doing so, use the internet as a tool of empowerment [e.g. Margolis, Resnick, & Levy, 2003]). This evidence is especially clear in the case of GroenLinks, which was, at the time of the campaign, one of the smallest parties in parliament (now the fifth party). Furthermore, we provide an insightful point of view into Anstead's question of whether "parties develop data-driven capabilities more rapidly in electoral systems with a tendency towards disproportionate outcomes" (2017; p. 23). In comparison with Hersh (2015), we focus less on how differences in data-availability lead to different strategic decisions, and more on how differences in the perception of campaign level and system level factors lead to a variation in the occurrence in PBT-innovation. With regard to Kreiss (2016), we have extended his model and applied it to a multiparty democracy. In comparison with Nielsen (2012), we focus solely on the perception of campaign leaders and not on canvassers. Furthermore, we focus on PBT on online as well as offline platforms.

Our attention for the system level factors has enabled us to identify perceived influence of the PR-system on the adoption of PBT. Contrary to theoretical expectations (Plasser & Plasser, 2002), campaigns in a one-district PR-system do identify key-areas that are more heavily campaigned than other districts. These key-areas differ from 'battleground states' in FPTP-systems in the sense that the key-area does not sometimes swing one way and sometimes the other, but rather that potential voters in key-areas are supportive of a certain party, but not very likely to show up at the polls. Campaigns use PBT-techniques to convince these potential voters of the personal relevance of politics and to motivate them to cast their vote. Areas with firm turnout numbers and clear support for a certain party, in contrast, are perceived as less decisive and less of a priority. This leads to a hierarchy of areas, which differs per party. Also, as a PR-system typically leads to a relatively large number of parties partaking in an election, PBT can be seen as an asset for a campaign to organise in a more efficient manner. Moreover, according to the campaign leaders, PBT-techniques offered by Facebook do allow smaller parties a degree of visibility that they are unable to achieve through traditional media.

On a campaign level, in the coming years, we expect more citizen-initiated campaigning (Gibson, 2015) by campaigns low in labour-resources. This requires a solid infrastructure, which opens the door for third party intermediaries offering off-the-shelf infrastructure. In this regard, it would be interesting to track the development of PvdA, which has suffered its biggest loss in history. This critical event could lead to the prototyping (Kreiss, 2016) of GroenLinks' innovative campaign by PvdA. As the party's chairman has resigned, the door is open to a more cognitive diverse party structure (Du Pre, 2017; De Vaan et al., 2015). Of course, these developments might apply less to parties that are officially more cognisant of campaign ethics (e.g. D66). This is why ethics and legal aspects are important factors to take into consideration. It would be interesting to see how these campaigns act as PBT-capabilities of rival parties improve. Their self-imposed barrier can limit their future chances, but can also attract voters growing more aware of the value of privacy. In the former case, this could lead to an overhaul of their privacy principles, or perhaps to a legislative push towards the restriction of PBT (similar to Hersh, 2015). In the latter case, campaigns can be expected to develop innovative non privacy-invasive campaigning techniques. Either way, our model would provide tools to study the process.

So 'can political parties campaign in Europe as they do in North America' (Bennett, 2016)? We would say ‘mostly yes’. We agree with Bennett (2015) that there are important differences between the US and Europe, and indeed, they influence how PBT is used. But based on our findings, we are hesitant to conclude that those differences (severely) constrain the export of PBT-practices to European multiparty systems. We have shown that relatively small campaign budgets do not need to bar parties from engaging in PBT-practices (or even from cooperating with BSD, an 'expensive' American political consultancy). The same is true of the electoral system: campaign leaders generally perceive PBT-techniques as useful in a PR-system. What remains is the relatively strict Dutch data protection law, labelling political preference as 'sensitive personal data', which can only be processed with explicit consent from the potential voter. 'Explicit consent', however, sounds harsher on paper than it is in practice and is easily achieved (e.g., Beales & Muris, 2008; Calo, 2012; Joergensen, 2014). Of course, because of data regulations and/or their non-existence, European campaigns are unable to consult voting lists showing whether an individual showed up at the polls in the last elections. In most European countries, the electoral register is inaccessible to political parties. One might argue that, from a campaign's perspective, US voter data are superior to European voter data. We would argue that European data are different, but they do not bar European campaigns in the use of PBT-techniques. Dutch campaigns, for instance, can (and do) rely on election results on voting booth level (which comprises a couple of streets). They can (and do) combine these results with detailed, accurate, and a multitude of data about the neighbourhoods surrounding those voting booths. And then there is Facebook, facilitating easy targeting of its users with personalised messages. As potential challenges for democracy come with PBT, such as ignoring 'less valuable' citizens (e.g. reliable non-voters), more research into the workings and effects of PBT is needed.

Disclaimer: This guest essay in the Special issue on political micro-targeting has not been peer reviewed. It is treated here as a reflection.

During the past three decades there has been a persistent, and dark, narrative about political micro-targeting. Phil Howard (2006) vividly described a present and future where politicians would use data to “redline” the citizens that received political information, manufacturing attitudes and beliefs, leading to “managed citizenship”. In the years since Howard wrote his monumental book, the concerns over micro-targeting have only grown. The explosion of data about the electorate in Western democracies such as Australia, Canada, the UK, and the United States (Howard & Kreiss, 2010) has triggered deep unease among scholars and privacy advocates alike. Sophisticated voter databases now contain everything from political party data gleaned through millions of interactions with the electorate, public data obtained from state agencies, and commercial marketing information that is bought and sold on international open markets. The 2016 US presidential election revealed the new ways that individuals can be profiled, identified, found, tracked, and messaged to on social media platforms such as Facebook and YouTube, which these companies themselves help facilitate (Kreiss and McGregor, 2017).

While it might seem that the micro-targeting practices of campaigns have massive, and un-democratic, electoral effects, decades of work in political communication should give us pause. Although we lack the first-hand data from political campaigns, consultancies, and technology firms such as Facebook to know for sure, previous research tells us that people are seldom the unwitting dupes of strategic political communication. Partisanship shapes much of how people vote and decades of research reveals that it is really hard to change people’s minds through campaigns (Kalla & Broockman, 2017; Henderson & Theodoridis, 2017). This has large implications for the effectiveness of micro-targeting. For example, Eitan Hersh’s (2015) deeply and carefully researched ground-breaking study using data from a major vendor to the US Democratic Party discovers that campaign practitioners find it really hard to persuade voters. This is because political practitioners lack reliable and identifiable data on cross-pressured and low information voters. Given this, campaigns often focus on known voters rather than risk targeting and messaging to the wrong people. Indeed, Hersh reveals that despite hundreds of data points on members of the electorate, it is a small cluster of publically available data – such as turnout history, party identification, and demographic data – that matters far more for predicting vote choice.

The lesson is that micro-targeted campaign ads are likely most effective in the short run when campaigns use them to mobilise identified supporters or partisans, spurring volunteerism, donations, and ultimately turnout – hardly the image of a managed, manipulated, or duped public (Baldwin-Philippi, 2017). Ironically, campaigns often use micro-targeting to further these forms of democratic participation, making appeals to targeted subsets of voters on the basis of the parties and issues they already care about. Campaigns also use micro-targeting in the attempt to decrease voter turnout on the opposing side, sending negative messages to the oppositions’ likely voters in the hopes this will make them less excited to turn out for their candidate. But two decades of social science suggests that this can be a risky strategy given that partisans can rally behind their candidate who is being attacked (Dunn & Tedesco, 2017).

What explains the outsized concerns about micro-targeting in the face of the generally thin evidence of its widespread and pernicious effects? This essay argues that we have anxieties about micro-targeting because we have anxieties about democracy itself. Or, to put it differently, that scholars often hold up an idealised vision of democracy as the standard upon which to judge all political communication. In a world where many scholars and journalists both hope and ardently believe, in the face of all available evidence, that members of the public are fundamentally rational, seek to be informed, and consider the general interest, micro-targeting appears to be manipulative, perverting the capacity of citizens to reason about politics. Meanwhile, for many scholars and journalists, political elites are fundamentally opposed to members of the public, seeking domination or control as opposed to representing their interests. In this world, much of the concern over micro-targeting reads as a classic “third-person effect”, where scholars and journalists presume that members of the public are more affected by campaign advertising than they themselves are.

And yet, this idealised version is not how democracy really is, nor necessarily how it should be. The argument of this brief essay is that, as a quantifiable practice premised on strategically identifying targeted groups of voters and crafting messages designed to appeal to them, micro-targeting is broadly reflective of the fact that democracy is often partisan, identity-based, and agonistic – in short, political. Following communication scholar Michael Schudson’s (1986) study of commercial advertising nearly three decades ago, this essay asks the following questions in the US context: what is the work that micro-targeting does, where does it fit into the political culture, and, what kind of political culture has given rise to it? I argue that micro-targeting is only imaginable, and efficacious, in a polity that prizes partisan mobilisation, group solidarity, agonism, and the clash of opposing moral views in its politics. Following from this, I suggest different democratic concerns about micro-targeting that relate to its cultural power to, over time, create a powerful set of representations of democracy that undermines the legitimacy of political representation, pluralism, and political leadership.

The cultural work of micro-targeting

To analyse the role that micro-targeting plays in politics, first we need to understand how and why citizens vote. In their recent book Democracy for Realists, political scientists Christopher Achen and Larry Bartels (2016) offer a sustained critique of what they call the “folk theory” of American democracy. According to this “folk theory” that underlies conceptions of popular sovereignty, Americans have identifiable and consistent policy preferences. During the course of an election, they inform themselves about the policy positions of candidates and make rational decisions as to which best represents their preferences, which in turn leads parties to be responsive to the wishes of the public.

As Achen and Bartels (ibid.) argue, this is a fiction. They outline a “group theory of democracy”, where it is social attachments and group identification that largely determine both partisanship and vote choice. Achen and Bartels argue that people see themselves in relation to the groups that they belong to and those that they do not. Identity is so strong, in this account, that it conditions both what partisans believe parties stand for but also their interpretation of facts (ibid., 267; see also Prasad et al., 2009). As Achen and Bartels demonstrate, this identity and group theory of politics has expansive empirical support over seventy years of research which demonstrates, time and again, that people have little knowledge about politics and yet detailed understandings of the social groups that the Democratic and Republican parties are perceived to represent. It is in this context that candidate performances of partisan and social identity become more important for electoral outcomes then the informational content of journalism. Events and candidates make identity more or less salient and strengthen group attachments. During campaigns, parties and candidates work to remind voters of their partisan and social attachments and strengthen them so they are mobilised to participate in the election. As Achen and Bartels (ibid., 311) argue:

Political campaigns consist in large part of reminding voters of their partisan identities – “mobilizing” them to support their group at the polls. Formal communications by the groups and informal communication networks among group members also help citizens understand how their identity groups connect to the candidates and parties.

In this context, what is important about political campaigns is this work of communicating the partisan and social identities of candidates to voters. Candidates and their campaigns use micro-targeting, along with other strategic communications, to accomplish this. Micro-targeting is both a campaign practice of using data to craft and deliver strategic messages to subsets of the electorate (historically across many different media), and a genre of campaign communications that, much like political advertising more broadly, reinforces and amplifies the partisan, group, and identity conflicts at the heart of US politics. There has been extensive research on how micro-targeting works as a data-driven and quantifiable practice (see, for instance, Karpf, 2016). What these messages do as a genre of campaign communications, however, has received considerably less scrutiny. Drawing on my own previous work in the US context (Kreiss, 2016), the first argument that I develop here is that micro-targeting furthers the mobilisation that Achen and Bartels (2015) identify, primarily through reminding citizens of and shoring up their partisan and group identities. I then discuss the potential democratic consequences of this in a more expansive, cultural sense.

Micro-targeted ads have an aesthetic of what I call “political realism”, building on Michael Schudson’s work on commercial advertising. In Advertising, The Uneasy Persuasion, Schudson (1986) compared commercial advertising with Soviet realist art (the official state-sanctioned art of the former Soviet Union), arguing that it offers a form of “commercial realism”. As commercial realism, commercial advertising “simplifies and typifies” (215); advertising is abstracted, presenting the world as it should be, not as it is, and it exemplifies individuals as members of larger social groups. As it does so, “the aesthetic of capitalist realism — without a masterplan of purposes — glorifies the pleasures and freedoms of consumer choice in defense of the virtues of private life and material ambitions.” (ibid., 218) 

We can see micro-targeted digital advertising as a cultural form of ‘political realism’ that reflects, reinforces, and celebrates a political culture, at least in the United States, premised on identity, moral certainty, and mobilisation - not weighty considerations of the general interest or deliberation. Micro-targeted digital content shares a few central characteristics, which I adapt here for politics from Schudson’s (1986) work on commercial realism:

• It presents social and political life in simplified and typified ways;
• It presents life as it should become, or for negative ads, as it must not become;
• It presents reality in its larger social significance, not in its actual workings;
• It presents progress towards the future and positive social struggle, or for negative ads, the ideas of the other party as negative steps back into the past. It carries a message of optimism for one partisan side, and takes a stance of pessimism towards political opponents; and,
• It tells us that political conflict is necessary, a clash of different groups and worldviews; moral certainty is assured, political identity is certain, and political agonism is reality.

For example, micro-targeted ads present social life in simplified ways, not presenting actual lives but abstract, stylised ones designed to be rife with larger meaning. A depiction of a farmer’s daily work in a campaign ad, for instance, is not about actual events or daily labours, but is meant to be an abstract, simplified, symbol of the American values of hard work and cultivation of the earth and celebration of ordinary people in a democratic society. The farmer here is typified; the campaign ad is not about a real person who farms. The farmer is a representation of the larger social categories, values, and ideas the ad presents as desirable or worthy of emulation for all Americans. At the same time, the two dominant US political parties often stress different themes in their ads, a recognition that they have different visions of what life should be become, what progress is, and what worldviews and moral claims the public should embrace. While doing so, political micro-targeting is inherently pluralist. It reflects a basic claim that “everyone has interests to defend and opinions to advance about his or her own good, or the group’s good, or the public good, and every interest was at least potentially a political interest group.” (Rosenblum, 2010, 259)

While it is impossible to know the full range of micro-targeted ads run during the course of an election cycle, consider some of the examples culled from the non-profit and non-partisan Democracy in Action website that chronicles US campaigns and the Hillary for America Design 2016 website that compiles the creative design from the campaign. To start, much of political micro-targeting is about building campaign databases by finding supporters online, signing them up for the cause through email, and repeatedly messaging them to enlist them in becoming a volunteer or a donor.

Take, for instance, the declarative “I am a Hillary Voter” digital ad (see Figure 1), presumably (but also logically) directed (although we can never know for sure) at the candidate’s supporters. What separates micro-targeted political ads from their mass broadcast counterparts is the data that lies behind them: campaigns can explicitly try to find and send messages to their partisan audiences or intra-party supporters, linking the names in their databases to identities online or on social media platforms such as Facebook. Campaigns can also try to find additional partisans and supporters by starting with the online behaviours, lifestyles, or likes or dislikes of known audiences and then seeking out ‘look-alike audiences’, to use industry parlance. And, what people do when they see these ads is quantified in terms of their performance, measured through things such as engagement and click-throughs. Micro-targeting is about mobilisation through conveying and building social solidarity. While there is much concern over candidates speaking out of both sides of their mouths to the electorate through hyper-targeted digital ads, likely far more often campaigns use micro-targeting to provide occasions for social identification and group belonging, conveying and constructing the sense of shared identity and group membership at the heart of politics. The “Wish Hillary a Happy Birthday” ad captures this (see Figure 2). Not only is this appeal directed at supporters (what Republican will want to wish Hillary a happy birthday after all), it constructs a sense of what social identification with Hillary Clinton means: motherhood, family, warmth, care, and nurturing.

Figure 1: Hillary Clinton digital campaign advertisements

Figure 2: Hillary Clinton digital campaign advertisement

Source: Hillary for America Design 2016

Micro-targeting is also about the marking of difference. This is, perhaps, the most common trope in micro-targeted digital campaign ads. Campaigns look to not only establish the cultural meaning of their candidates and supporters, but also that of their opposition (Alexander, 2010). Donald Trump’s ads during the 2016 election reflected his rhetoric from the campaign trail in stressing themes of safety and security, in addition to the need to draw boundaries around civic incorporation (i.e., who should be allowed to be a citizen). For Hillary Clinton, micro-targeted ads were celebrations of diversity and multi-culturalism, especially the empowerment of women and racial and ethnic minorities. Political advertisements attempt to connect the candidates they promote with the demographic and social groups they seek to represent (in the United States this is at times drawn on racial and ethnic terms: whites for Republicans and a more diverse coalition for Democrats, see the discussion in Grossmann & Hopkins, 2016, 43-45).

In this, micro-targeting reflects and reinforces political agonism, the clash of competing social groups, interests, and values. Through micro-targeting, candidates stake out their claim to be on the civil side of the moral binary of the political sphere and strive to paint their opponents as anti-civil (Alexander, 2010). More colloquially, micro-targeted advertisements offer the beautiful affirmation of our values and the sharp critique of those of our opponents. Hillary Clinton’s campaign, for instance, clearly sought to portray Trump in terms of anti-civil racism, xenophobia, and sexism. And, the campaign used issues, such as abortion rights, and values, such as autonomy and choice, to build group identity and social solidarity around opposition to Trump: “Let’s stand together, join millions of women” (see Figure 3). This Facebook ad pits Clinton and her supporters against Trump and his supporters. Trump, in turn, combined nationalist and security appeals with an implicit construction of the American body politic in white identity terms (Figure 4). These ads capture the reality that political conflict is not only inevitable, but necessary: there are opposing views in politics on fundamental questions such as life, autonomy, and country. The audiences for these ads are not being presented with information to help them make up their own minds, they are being invited into a political struggle with clear opposing worldviews and moral values (see Figure 5). This is why mobilisation ads are directed towards identity-congruent audiences.

Figure 3: Hillary Clinton Facebook advertisement

Figure 4: Donald Trump digital advertisement

Source: Democracy in Action

Figure 5: Anti-Trump Hillary Clinton digital advertisements

Source: Hillary for America Design 2016

In these advertisements, it is also clear that micro-targeted ads present life as it should become, or as it must not become, linking the preferred candidate and political party with a civil vision of the future and the opposition with an anti-civil vision of the future, to use Alexander’s (2010) framework. As an example, for Ted Cruz (see Figure 6), the opposing side wants to infringe on the Bill of Rights, the fundamental liberty of Americans to defend their lives, liberties, families, and properties. Candidates run these issue ads to stake out their stance on the conflicting values, visions of the good life, plans for the future, and ends that are desirable in politics – whether it is embracing the freedom and security of gun rights for American Republicans or autonomy and choice in the context of reproductive rights for Democrats. These appeals are designed to mobilise the committed around the candidate’s vision of America’s past and future – they are designed for a world where we are sure of who we are and committed to our values and the ends we pursue.

Figure 6: Ted Cruz digital campaign advertisement

Source: Democracy in Action

Conclusion: democratic anxieties

I believe that there is such democratic anxiety about micro-targeting because citizens are supposed to be independent, autonomous, and rational. Micro-targeted advertising works to reinforce group identities and solidarity, mobilise partisans, and further the clash of political values. These things are all suspect from the perspective of the powerful and potent “folk theory” of democracy, as Achen and Bartels phrase it. As these realists argue, however, it’s far better to grapple with the reality of group-based democracy, with its attendant ingrained social allegiances and conflicts over values and power, rather than wishing for a transcendent and pure form of democracy without politics. These authors argue that we need to make peace with conflictual and competitive forms of group-based and pluralistic democracy premised on institutionally organised opposition. As Achen and Bartels (2015, 318) conclude:

Freedom is to faction what air is to fire, Madison said. But ordinary citizens often dislike the conflict and bickering that comes with freedom. They wish their elected officials would just do the people’s work without so much squabbling amongst themselves. They dislike the compromises that result when many different groups are free to propose alternative policies, leaving politicians to adjust their differences. Voters want “a real leader, not a politician,” by which they generally mean that their own ideas should be adopted and other people’s opinions disregarded, because views different from their own are obviously self-interested and erroneous. To the contrary, politicians with vision who are also skilled at creative compromise are the soul of successful democracy, and they exemplify real leadership.

My own view is that micro-targeting comes in the necessary service of this “conflict and bickering”. At its normative best, micro-targeting strengthens the hands of opposing factions, enabling them to identify and mobilise partisans to their cause, providing them with resources in terms of boots on the ground and money in the coffers. When opposing politicians and parties square off, they carry these resources into battle trying to advance their agendas or win concessions for their side. Compromise may be harder in a world of stronger factions, their hands steadied by the resources that micro-targeting can deliver, but that does not make compromise any less necessary or essential.

On the other hand, there are reasons for democratic concern about micro-targeting, but they look a bit different from narratives about public manipulation. Schudson (1986, 232) concludes that “advertising does not make people believe in capitalist institutions or even in consumer values, but so long as alternative articulations of values are relatively hard to locate in the culture, capitalist realist art will have some power.” I suspect that the same is true of political micro-targeting. The cultural power of political micro-targeting, but also political advertising more generally, lies in its creation of a set of ready-to-hand representations of democracy that citizens can express easily and fall back on. Taken to its extreme in a polarized political climate, micro-targeting can work to undermine the legitimacy of conflicts over opposing values and claims in democratic life. For example, in an undemocratic political culture micro-targeting can portray the other side as crooked and dangerous to the polity, political compromise as selling out, political expertise and representation as not to be trusted, and partisans’ own beliefs and identities as the only legitimate ones, not simply those among many in a pluralistic democracy. Micro-targeting also melds symbolic and social power in new ways, culturally legitimating and furthering the fortunes of autonomous and independent candidates, divorced from their parties and taking their appeals directly to voters (see Hersh, 2017).

Disclaimer: This guest essay in the Special issue on political micro-targeting has not been peer reviewed. It is an abbreviated version of a speech delivered by the Member of the European Partiament (MEP) Sophie in ‘t Veld in Amsterdam in May 2017 to Data & Democracy, a conference on political micro-targeting.

Democracy

Democracy is valuable and vulnerable, which is reason enough to remain alert for new developments that can undermine her. In recent months, we have seen enough examples of the growing impact of personal data in campaigns and elections. It is important and urgent for us to publicly debate this development. It is easy to see why we should take action against extremist propaganda of hatemongers aiming to recruit young people for violent acts. But we euphemistically speak of 'fake news' when lies, 'half-truths’, conspiracy theories, and sedition creepily poison public opinion.

The literal meaning of democracy is 'the power of the people'. 'Power' presupposes freedom. Freedom to choose and to decide. Freedom from coercion and pressure. Freedom from manipulation. 'Power' also presupposes knowledge. Knowledge of all facts, aspects, and options. And knowing how to balance them against each other. When freedom and knowledge are restricted, there can be no power.

In a democracy, every individual choice influences society as a whole. Therefore, the common interest is served with everyone's ability to make their choices in complete freedom, and with complete knowledge.

The interests of parties and political candidates who compete for citizen’s votes may differ from that higher interest. They want citizens to see their political advertising, and only theirs, not that of their competitors. Not only do parties and candidates compete for the voter's favour. They contend for his exclusive time and attention as well.

Political targeting

No laws dictate what kind of information a voter should rely on to be able to make the right consideration. For lamb chops, toothpaste, mortgages or cars, for example, it’s mandatory for producers to mention the origin and properties. This enables consumers to make a responsible decision. Providing false information is illegal. All ingredients, properties, and risks have to be mentioned on the label.

Political communication, however, is protected by freedom of speech. Political parties are allowed to use all kinds of sales tricks.

And, of course, campaigns do their utmost and continuously test the limits of the socially acceptable.

Nothing new, so far. There is no holding back in getting the voters to cast their vote on your party or your candidate. From temptation with attractive promises, to outright bribery. From applying pressure to straightforward intimidation.

Important therein is how and where you can reach the voter. In the old days it was easy: Catholics were told on Sundays in church that they had no other choice in the voting booth than the catholic choice. And no righteous Catholic dared to think about voting differently. At home, the father told the mother how to vote. The children received their political preference from home and from school. Catholics learned about current affairs via a catholic newspaper, and through the catholic radio broadcaster. In the Dutch society, which consisted of a few of such pillars, one was only offered the opinions of one's own pillar1. A kind of filter bubble avant la lettre.

Political micro-targeting

Nowadays, political parties have a different approach. With new technologies, the sky is the limit.

Increasingly advanced techniques allow the mapping of voter preferences, activities, and connections. Using endless amounts of personal data, any individual on earth can be reconstructed in detail. Not only can their personal beliefs be distilled from large troves of data, no, it even is possible to predict a person's beliefs, even before they have formed them themselves. And, subsequently, it is possible to subtly steer those beliefs, while leaving the person thinking they made their decision all by themselves.

As often is the case, the Americans lead in the use of new techniques. While we Europeans, touchingly old-fashioned knock on doors and hand out flyers at Saturday's market, the American employ the latest technology to identify, approach, and influence voters.

Of course, trying to find out where voters can be reached and how they can be influenced is no novelty. Political parties map which neighbourhoods predominantly vote for them, which neighbourhoods have potential, and in which neighbourhoods campaigning would be a wasted effort. Parties work with detailed profiles and target audiences, for which they can tailor their messages.

But the usage of personal data on a large scale has a lot more to offer. Obviously, this is a big opportunity for political parties, and for anyone else, who runs campaigns or aims to influence the elections.

However, the influencing techniques become increasingly opaque. As a result of the alleged filter bubble, voters are being reaffirmed in their own beliefs, and they hardly receive information anymore about the beliefs and arguments of other groups. This new kind of segmentation may stifle critical thinking. There may not be enough incentive to test one's own ideas, to find new arguments, or to critically reflect on the truthfulness of information.

I am a social and economic liberal D66 politician, and I get suggestions for news articles from websites like The Guardian or Le Monde. My colleague from the right wing nationalist PVV, may well receive URLs from Breitbart.

Pluralism is essential for a healthy, robust democracy. In a polarised society, people live in tightly knit groups, which hardly communicate with each other. In a pluralist society people engage in the free exchange, confrontation, and fusion of ideas.

The concept pluralism is under pressure. Populist parties declare themselves representative of The People. In their vision, The People, is uniform and homogenous. There is a dominant cultural norm, dictated from the top-down, to which everyone must conform. Whomever refuses, gets chewed out. Often, it is about one-dimensional symbolism such as Easter eggs and Christmas trees. There is no place for pluralism in the world of the populists. But when there is no pluralism, there is no democracy. Without pluralism, democracy is nothing more than a simple tribal dispute, instead of the expression of the will of all citizens together.

Voter data

European privacy legislation limits the use of personal data. In the world of ‘big data’, one of the explicit goals of regulation is to prevent restriction of the consumer's choice. Oddly enough, lawmakers do not explicitly aspire to guarantee voters as broad a choice as possible. But in politics, individual choices have consequences for society as a whole.

In 2018, the General Data Protection Regulation (GDPR) comes into effect. We have worked five years on the GDPR. At this moment, we work on the modernisation of the e-Privacy Directive, which is mainly about the protection of communication. As was the case with the GDPR, companies from certain sectors scream bloody murder. European privacy protection would mean certain death for the European industry. According to some corporate Cassandras, entire European industries will move to other continents. That very same death of corporate Europe is also predicted for any measure concerning, say, environmental norms, procurement rules, or employee rights. All those measures are in place, but, as far as I know, the nightmare scenario has never occurred...

There are some corporate sectors, such as publishing and marketing, which have a huge impact on the information supply to citizens. They are the ones who now cry wolf. It is understandable that they are unhappy with stricter rules concerning their activities, but as the potential impact of the use of personal data and ‘big data’ increases, so does their social responsibility.

At the moment, there is not much public debate about the new techniques. Peculiar. Thirty years ago, 'subliminal advertising', as we called it then, was prohibited because people found it unethical to influence people without their knowledge. We need to have a similar debate. What do we think of opaque influencing? Do we need ethical norms? Should such norms apply only to political campaigns, or should we look at this from a broader perspective? In the ‘big data’ debate, we tend to speak in technical or legal terms, while actually the issue is fundamentally ethical, holding far-reaching consequences for the vitality of our democracy.

Such a public debate demands more clarity on the impact of ‘big data’, profiling, targeting, and similar techniques on the individual, her behaviour, and her choices, which determine in what direction society progresses. Which voters are being reached? How sensitive are they for the subtle influencing and what makes them resilient? How do people who are hardly reached only compare to the others? How do voters and non-voters compare? Is the voter truly predictable? Can we identify or influence the floating voter? Do voters actually float between different parties? Or do they especially float within their own party, their own bubble, their own segment? How important are other factors, such as the social context? If the new influencing techniques are indeed as potent as we think, how can polls get it so wrong? What can we learn from advertisers who return to contextual advertising, because targeting turns out less effective than they thought?

We need to stay cool-headed. New technologies have a huge impact, but human nature will not suddenly change due to ‘big data’ and its use. Our natural instincts and reflexes will definitely not evolve in a few years. That would take many thousands of years, as even in the 21st century, we seem to have more than a few cavemen traits, so losing internalised behaviour is not as easy as 1-2-3. Humans are resilient, but democracy is vulnerable. On a short term, the societal impact is large. This gives us all the reason to reflect on how to deal with the new reality, and how we can keep up our values in this new reality.

The use of personal data, clearly, is not solely reserved for decent political parties. Other persons and organisations, from the Kremlin to Breitbart, can bombard European voters with information and misinformation. But European governments, controlling endless amounts of personal data of their citizens, can also manipulate information, or circulate utter nonsense to advance their own interests. A random example: the Hungarian government influencing their voters with lies and manipulation about the so-called consultation on asylum seekers.

Beyond voter data

This issue is not only about the personal data of voters, but also about the personal data of political competitors, opponents, and critics, which are increasingly being employed. Recently, we have seen efforts of external parties to influence the results of the 2017 French elections. We saw a large-scale hack of the Emmanuel Macron campaign, and the spread of false information, coming obviously from the Kremlin and the American Alt-Right, meant to discredit Macron's candidacy.

Also, the American elections show the shady game of hacking, leaking, and manipulating. The issue of the Hillary Clinton mails will undoubtedly occupy our minds for years. Who knows how the elections would have turned out without this affair?

Other democratic pillars can get corrupted as well by the misuse of data. Critical voices, opposition, and checks and balances are democracy's oxygen. Democracy is in acute jeopardy when data are employed to attack, undermine, discredit, blackmail, or persecute journalists, judges, lawyers, NGOs, whistleblowers, and opposition parties.

In Europe, we tend to shrug our shoulders at these dangers. "Oh well, we'll see, such things occur only in banana republics, not right here". Of course, this trust in our democratic rule of law is wonderful. But if we treat our rule of law this neglectfully, we will lose it eventually.

Within the European Union, we currently see this happening in Poland and Hungary. The governments of both nations ruthlessly attack independent judges, critical media, inconvenient NGOs. They do so with quasi-lawful means. Under the banner of transparency, they force NGOs to register. In doing so, they misuse laws against money laundering, and terror finance. Or the governments bring out compromising information about judges or politicians in strategic moments.

But critical voices struggle in other member states as well. Lawyers are being monitored, even without a legal basis. In the years after 9/11, we have created endless new abilities for intelligence services, police and justice departments to spy on citizens, even without suspicion, without the signature of a judge. The companies to which we unwittingly surrender our personal data, in exchange for service, are forced to hand over all information to the government, or forced to build in backdoors. Governments hack computers in other countries. Usually, it starts out with unlawful practices, but soon enough laws are put in place to legalise those practices. The magic word 'terrorism' silences any critique on such legislation.

But when politicians, journalists, NGOs, whistleblowers, lawyers, and many others cannot perform their tasks freely and without worry, our democracy withers. Not only do they have to operate without someone keeping an eye on them, they have to know nobody is in fact watching them. The mere possibility of being watched, results in a chilling effect.

For this principal reason, I have contested a French mass surveillance law before the French Conseil d'Etat. Since, as a member of the European Parliament, I spend four days a month on French soil (in Strasbourg), I could potentially be the target of the French eavesdropping programme. This is not totally imaginary, as I am not only a politician, but also a vocal critic of certain French anti-terror measures. It is not about me actually worrying about being spied on, but about the fact that I might be spied on. Luckily, I am not easily startled, but I can imagine that many politicians are vulnerable. That is a risk for democracy.

I do not discard the possibility of a ruling of the European Court of Human Rights on my case. In that turn of events, it will lead to jurisprudence valid in the entire EU (and the geographical area covered by the Council of Europe).

But, of course, this should not depend on the actions of one obstinate individual whether politicians, NGOs, journalists, and so on, can do their jobs fearlessly, to fulfil their watchdog role.

It is my personal, deep, conviction that the biggest threat to our democracy is the fact that we have enabled the powerful to access, with almost no limitations, the personal data of those who should control those very same powerful entities.

What can we do?

Some propose new forms of democracy, in which universal suffrage is weakened or even abolished. In his book ‘Against elections: the case for democracy’, David Van Reybrouck had the idea to appoint representatives on the basis of chance, and in his book ‘Against democracy’ Jason Brennan wants to give the elite more votes than the lower classes, presuming that people with more education or development make better choices. Others want to replace representative democracy with direct democracy.

I oppose those ideas. Universal suffrage and the representative democracy are great achievements, which have led to enormous progress in society.

First of all, we have to make sure our children grow up to be critical, independent thinkers. Think differently, deviate, provoke: this must be encouraged instead of condemned. A democracy needs non-conformists.

We must teach our children to contextualise information and to compare sources.

The counterpart of ‘big data’ must be ‘big transparency’. We need to understand not just open administration, but also insights into the techniques of influence.

The regulation and limitation of the use of personal data, as I hope to have argued effectively, is not a game of out-of-touch privacy activists. It is essential for democracy. We need safeguards, not only to be sure people really are free in their choices, but also to protect the necessary checks and balances. As such, I plea for a rigorous application of the GDPR, and in the European Parliament, I will work for a firm e-Privacy Directive.

And yes, perhaps we should examine whether the rules for political campaigning are still up-to-date. In most countries, those rules cover a cap on campaign expenditures, a prohibition of campaigning or polling on the day before election day, or a ban on publishing information that may influence the election results, such as the leaked e-mails in France. But these rules have little impact on the use of personal data to subtly influence elections.

Last year, the European Parliament supported my proposal for a mechanism to guard democracy, the rule of law, and fundamental rights in Europe.2

On this day (editor’s note: 9 May, Europe Day) of European democracy, I plead for equal, high norms in Europe. The last years have shown that national elections are European elections. It is crucial for us to trust that all elections in EU member states are open, free, and honest elections, free of improper influencing.

These last sixty years, the European Union has developed itself into a world leader in democracy and freedom. If we start a public debate, Europe can remain a world leader.

Disclaimer: This guest essay in the Special issue on political micro-targeting has not been peer reviewed. It is treated here as a reflection.

During the past three decades there has been a persistent, and dark, narrative about political micro-targeting. Phil Howard (2006) vividly described a present and future where politicians would use data to “redline” the citizens that received political information, manufacturing attitudes and beliefs, leading to “managed citizenship”. In the years since Howard wrote his monumental book, the concerns over micro-targeting have only grown. The explosion of data about the electorate in Western democracies such as Australia, Canada, the UK, and the United States (Howard & Kreiss, 2010) has triggered deep unease among scholars and privacy advocates alike. Sophisticated voter databases now contain everything from political party data gleaned through millions of interactions with the electorate, public data obtained from state agencies, and commercial marketing information that is bought and sold on international open markets. The 2016 US presidential election revealed the new ways that individuals can be profiled, identified, found, tracked, and messaged to on social media platforms such as Facebook and YouTube, which these companies themselves help facilitate (Kreiss and McGregor, 2017).

While it might seem that the micro-targeting practices of campaigns have massive, and un-democratic, electoral effects, decades of work in political communication should give us pause. Although we lack the first-hand data from political campaigns, consultancies, and technology firms such as Facebook to know for sure, previous research tells us that people are seldom the unwitting dupes of strategic political communication. Partisanship shapes much of how people vote and decades of research reveals that it is really hard to change people’s minds through campaigns (Kalla & Broockman, 2017; Henderson & Theodoridis, 2017). This has large implications for the effectiveness of micro-targeting. For example, Eitan Hersh’s (2015) deeply and carefully researched ground-breaking study using data from a major vendor to the US Democratic Party discovers that campaign practitioners find it really hard to persuade voters. This is because political practitioners lack reliable and identifiable data on cross-pressured and low information voters. Given this, campaigns often focus on known voters rather than risk targeting and messaging to the wrong people. Indeed, Hersh reveals that despite hundreds of data points on members of the electorate, it is a small cluster of publically available data – such as turnout history, party identification, and demographic data – that matters far more for predicting vote choice.

The lesson is that micro-targeted campaign ads are likely most effective in the short run when campaigns use them to mobilise identified supporters or partisans, spurring volunteerism, donations, and ultimately turnout – hardly the image of a managed, manipulated, or duped public (Baldwin-Philippi, 2017). Ironically, campaigns often use micro-targeting to further these forms of democratic participation, making appeals to targeted subsets of voters on the basis of the parties and issues they already care about. Campaigns also use micro-targeting in the attempt to decrease voter turnout on the opposing side, sending negative messages to the oppositions’ likely voters in the hopes this will make them less excited to turn out for their candidate. But two decades of social science suggests that this can be a risky strategy given that partisans can rally behind their candidate who is being attacked (Dunn & Tedesco, 2017).

What explains the outsized concerns about micro-targeting in the face of the generally thin evidence of its widespread and pernicious effects? This essay argues that we have anxieties about micro-targeting because we have anxieties about democracy itself. Or, to put it differently, that scholars often hold up an idealised vision of democracy as the standard upon which to judge all political communication. In a world where many scholars and journalists both hope and ardently believe, in the face of all available evidence, that members of the public are fundamentally rational, seek to be informed, and consider the general interest, micro-targeting appears to be manipulative, perverting the capacity of citizens to reason about politics. Meanwhile, for many scholars and journalists, political elites are fundamentally opposed to members of the public, seeking domination or control as opposed to representing their interests. In this world, much of the concern over micro-targeting reads as a classic “third-person effect”, where scholars and journalists presume that members of the public are more affected by campaign advertising than they themselves are.

And yet, this idealised version is not how democracy really is, nor necessarily how it should be. The argument of this brief essay is that, as a quantifiable practice premised on strategically identifying targeted groups of voters and crafting messages designed to appeal to them, micro-targeting is broadly reflective of the fact that democracy is often partisan, identity-based, and agonistic – in short, political. Following communication scholar Michael Schudson’s (1986) study of commercial advertising nearly three decades ago, this essay asks the following questions in the US context: what is the work that micro-targeting does, where does it fit into the political culture, and, what kind of political culture has given rise to it? I argue that micro-targeting is only imaginable, and efficacious, in a polity that prizes partisan mobilisation, group solidarity, agonism, and the clash of opposing moral views in its politics. Following from this, I suggest different democratic concerns about micro-targeting that relate to its cultural power to, over time, create a powerful set of representations of democracy that undermines the legitimacy of political representation, pluralism, and political leadership.

The cultural work of micro-targeting

To analyse the role that micro-targeting plays in politics, first we need to understand how and why citizens vote. In their recent book Democracy for Realists, political scientists Christopher Achen and Larry Bartels (2016) offer a sustained critique of what they call the “folk theory” of American democracy. According to this “folk theory” that underlies conceptions of popular sovereignty, Americans have identifiable and consistent policy preferences. During the course of an election, they inform themselves about the policy positions of candidates and make rational decisions as to which best represents their preferences, which in turn leads parties to be responsive to the wishes of the public.

As Achen and Bartels (ibid.) argue, this is a fiction. They outline a “group theory of democracy”, where it is social attachments and group identification that largely determine both partisanship and vote choice. Achen and Bartels argue that people see themselves in relation to the groups that they belong to and those that they do not. Identity is so strong, in this account, that it conditions both what partisans believe parties stand for but also their interpretation of facts (ibid., 267; see also Prasad et al., 2009). As Achen and Bartels demonstrate, this identity and group theory of politics has expansive empirical support over seventy years of research which demonstrates, time and again, that people have little knowledge about politics and yet detailed understandings of the social groups that the Democratic and Republican parties are perceived to represent. It is in this context that candidate performances of partisan and social identity become more important for electoral outcomes then the informational content of journalism. Events and candidates make identity more or less salient and strengthen group attachments. During campaigns, parties and candidates work to remind voters of their partisan and social attachments and strengthen them so they are mobilised to participate in the election. As Achen and Bartels (ibid., 311) argue:

Political campaigns consist in large part of reminding voters of their partisan identities – “mobilizing” them to support their group at the polls. Formal communications by the groups and informal communication networks among group members also help citizens understand how their identity groups connect to the candidates and parties.

In this context, what is important about political campaigns is this work of communicating the partisan and social identities of candidates to voters. Candidates and their campaigns use micro-targeting, along with other strategic communications, to accomplish this. Micro-targeting is both a campaign practice of using data to craft and deliver strategic messages to subsets of the electorate (historically across many different media), and a genre of campaign communications that, much like political advertising more broadly, reinforces and amplifies the partisan, group, and identity conflicts at the heart of US politics. There has been extensive research on how micro-targeting works as a data-driven and quantifiable practice (see, for instance, Karpf, 2016). What these messages do as a genre of campaign communications, however, has received considerably less scrutiny. Drawing on my own previous work in the US context (Kreiss, 2016), the first argument that I develop here is that micro-targeting furthers the mobilisation that Achen and Bartels (2015) identify, primarily through reminding citizens of and shoring up their partisan and group identities. I then discuss the potential democratic consequences of this in a more expansive, cultural sense.

Micro-targeted ads have an aesthetic of what I call “political realism”, building on Michael Schudson’s work on commercial advertising. In Advertising, The Uneasy Persuasion, Schudson (1986) compared commercial advertising with Soviet realist art (the official state-sanctioned art of the former Soviet Union), arguing that it offers a form of “commercial realism”. As commercial realism, commercial advertising “simplifies and typifies” (215); advertising is abstracted, presenting the world as it should be, not as it is, and it exemplifies individuals as members of larger social groups. As it does so, “the aesthetic of capitalist realism — without a masterplan of purposes — glorifies the pleasures and freedoms of consumer choice in defense of the virtues of private life and material ambitions.” (ibid., 218) 

We can see micro-targeted digital advertising as a cultural form of ‘political realism’ that reflects, reinforces, and celebrates a political culture, at least in the United States, premised on identity, moral certainty, and mobilisation - not weighty considerations of the general interest or deliberation. Micro-targeted digital content shares a few central characteristics, which I adapt here for politics from Schudson’s (1986) work on commercial realism:

• It presents social and political life in simplified and typified ways;
• It presents life as it should become, or for negative ads, as it must not become;
• It presents reality in its larger social significance, not in its actual workings;
• It presents progress towards the future and positive social struggle, or for negative ads, the ideas of the other party as negative steps back into the past. It carries a message of optimism for one partisan side, and takes a stance of pessimism towards political opponents; and,
• It tells us that political conflict is necessary, a clash of different groups and worldviews; moral certainty is assured, political identity is certain, and political agonism is reality.

For example, micro-targeted ads present social life in simplified ways, not presenting actual lives but abstract, stylised ones designed to be rife with larger meaning. A depiction of a farmer’s daily work in a campaign ad, for instance, is not about actual events or daily labours, but is meant to be an abstract, simplified, symbol of the American values of hard work and cultivation of the earth and celebration of ordinary people in a democratic society. The farmer here is typified; the campaign ad is not about a real person who farms. The farmer is a representation of the larger social categories, values, and ideas the ad presents as desirable or worthy of emulation for all Americans. At the same time, the two dominant US political parties often stress different themes in their ads, a recognition that they have different visions of what life should be become, what progress is, and what worldviews and moral claims the public should embrace. While doing so, political micro-targeting is inherently pluralist. It reflects a basic claim that “everyone has interests to defend and opinions to advance about his or her own good, or the group’s good, or the public good, and every interest was at least potentially a political interest group.” (Rosenblum, 2010, 259)

While it is impossible to know the full range of micro-targeted ads run during the course of an election cycle, consider some of the examples culled from the non-profit and non-partisan Democracy in Action website that chronicles US campaigns and the Hillary for America Design 2016 website that compiles the creative design from the campaign. To start, much of political micro-targeting is about building campaign databases by finding supporters online, signing them up for the cause through email, and repeatedly messaging them to enlist them in becoming a volunteer or a donor.

Take, for instance, the declarative “I am a Hillary Voter” digital ad (see Figure 1), presumably (but also logically) directed (although we can never know for sure) at the candidate’s supporters. What separates micro-targeted political ads from their mass broadcast counterparts is the data that lies behind them: campaigns can explicitly try to find and send messages to their partisan audiences or intra-party supporters, linking the names in their databases to identities online or on social media platforms such as Facebook. Campaigns can also try to find additional partisans and supporters by starting with the online behaviours, lifestyles, or likes or dislikes of known audiences and then seeking out ‘look-alike audiences’, to use industry parlance. And, what people do when they see these ads is quantified in terms of their performance, measured through things such as engagement and click-throughs. Micro-targeting is about mobilisation through conveying and building social solidarity. While there is much concern over candidates speaking out of both sides of their mouths to the electorate through hyper-targeted digital ads, likely far more often campaigns use micro-targeting to provide occasions for social identification and group belonging, conveying and constructing the sense of shared identity and group membership at the heart of politics. The “Wish Hillary a Happy Birthday” ad captures this (see Figure 2). Not only is this appeal directed at supporters (what Republican will want to wish Hillary a happy birthday after all), it constructs a sense of what social identification with Hillary Clinton means: motherhood, family, warmth, care, and nurturing.

Figure 1: Hillary Clinton digital campaign advertisements

Figure 2: Hillary Clinton digital campaign advertisement

Source: Hillary for America Design 2016

Micro-targeting is also about the marking of difference. This is, perhaps, the most common trope in micro-targeted digital campaign ads. Campaigns look to not only establish the cultural meaning of their candidates and supporters, but also that of their opposition (Alexander, 2010). Donald Trump’s ads during the 2016 election reflected his rhetoric from the campaign trail in stressing themes of safety and security, in addition to the need to draw boundaries around civic incorporation (i.e., who should be allowed to be a citizen). For Hillary Clinton, micro-targeted ads were celebrations of diversity and multi-culturalism, especially the empowerment of women and racial and ethnic minorities. Political advertisements attempt to connect the candidates they promote with the demographic and social groups they seek to represent (in the United States this is at times drawn on racial and ethnic terms: whites for Republicans and a more diverse coalition for Democrats, see the discussion in Grossmann & Hopkins, 2016, 43-45).

In this, micro-targeting reflects and reinforces political agonism, the clash of competing social groups, interests, and values. Through micro-targeting, candidates stake out their claim to be on the civil side of the moral binary of the political sphere and strive to paint their opponents as anti-civil (Alexander, 2010). More colloquially, micro-targeted advertisements offer the beautiful affirmation of our values and the sharp critique of those of our opponents. Hillary Clinton’s campaign, for instance, clearly sought to portray Trump in terms of anti-civil racism, xenophobia, and sexism. And, the campaign used issues, such as abortion rights, and values, such as autonomy and choice, to build group identity and social solidarity around opposition to Trump: “Let’s stand together, join millions of women” (see Figure 3). This Facebook ad pits Clinton and her supporters against Trump and his supporters. Trump, in turn, combined nationalist and security appeals with an implicit construction of the American body politic in white identity terms (Figure 4). These ads capture the reality that political conflict is not only inevitable, but necessary: there are opposing views in politics on fundamental questions such as life, autonomy, and country. The audiences for these ads are not being presented with information to help them make up their own minds, they are being invited into a political struggle with clear opposing worldviews and moral values (see Figure 5). This is why mobilisation ads are directed towards identity-congruent audiences.

Figure 3: Hillary Clinton Facebook advertisement

Figure 4: Donald Trump digital advertisement

Source: Democracy in Action

Figure 5: Anti-Trump Hillary Clinton digital advertisements

Source: Hillary for America Design 2016

In these advertisements, it is also clear that micro-targeted ads present life as it should become, or as it must not become, linking the preferred candidate and political party with a civil vision of the future and the opposition with an anti-civil vision of the future, to use Alexander’s (2010) framework. As an example, for Ted Cruz (see Figure 6), the opposing side wants to infringe on the Bill of Rights, the fundamental liberty of Americans to defend their lives, liberties, families, and properties. Candidates run these issue ads to stake out their stance on the conflicting values, visions of the good life, plans for the future, and ends that are desirable in politics – whether it is embracing the freedom and security of gun rights for American Republicans or autonomy and choice in the context of reproductive rights for Democrats. These appeals are designed to mobilise the committed around the candidate’s vision of America’s past and future – they are designed for a world where we are sure of who we are and committed to our values and the ends we pursue.

Figure 6: Ted Cruz digital campaign advertisement

Source: Democracy in Action

Conclusion: democratic anxieties

I believe that there is such democratic anxiety about micro-targeting because citizens are supposed to be independent, autonomous, and rational. Micro-targeted advertising works to reinforce group identities and solidarity, mobilise partisans, and further the clash of political values. These things are all suspect from the perspective of the powerful and potent “folk theory” of democracy, as Achen and Bartels phrase it. As these realists argue, however, it’s far better to grapple with the reality of group-based democracy, with its attendant ingrained social allegiances and conflicts over values and power, rather than wishing for a transcendent and pure form of democracy without politics. These authors argue that we need to make peace with conflictual and competitive forms of group-based and pluralistic democracy premised on institutionally organised opposition. As Achen and Bartels (2015, 318) conclude:

Freedom is to faction what air is to fire, Madison said. But ordinary citizens often dislike the conflict and bickering that comes with freedom. They wish their elected officials would just do the people’s work without so much squabbling amongst themselves. They dislike the compromises that result when many different groups are free to propose alternative policies, leaving politicians to adjust their differences. Voters want “a real leader, not a politician,” by which they generally mean that their own ideas should be adopted and other people’s opinions disregarded, because views different from their own are obviously self-interested and erroneous. To the contrary, politicians with vision who are also skilled at creative compromise are the soul of successful democracy, and they exemplify real leadership.

My own view is that micro-targeting comes in the necessary service of this “conflict and bickering”. At its normative best, micro-targeting strengthens the hands of opposing factions, enabling them to identify and mobilise partisans to their cause, providing them with resources in terms of boots on the ground and money in the coffers. When opposing politicians and parties square off, they carry these resources into battle trying to advance their agendas or win concessions for their side. Compromise may be harder in a world of stronger factions, their hands steadied by the resources that micro-targeting can deliver, but that does not make compromise any less necessary or essential.

On the other hand, there are reasons for democratic concern about micro-targeting, but they look a bit different from narratives about public manipulation. Schudson (1986, 232) concludes that “advertising does not make people believe in capitalist institutions or even in consumer values, but so long as alternative articulations of values are relatively hard to locate in the culture, capitalist realist art will have some power.” I suspect that the same is true of political micro-targeting. The cultural power of political micro-targeting, but also political advertising more generally, lies in its creation of a set of ready-to-hand representations of democracy that citizens can express easily and fall back on. Taken to its extreme in a polarized political climate, micro-targeting can work to undermine the legitimacy of conflicts over opposing values and claims in democratic life. For example, in an undemocratic political culture micro-targeting can portray the other side as crooked and dangerous to the polity, political compromise as selling out, political expertise and representation as not to be trusted, and partisans’ own beliefs and identities as the only legitimate ones, not simply those among many in a pluralistic democracy. Micro-targeting also melds symbolic and social power in new ways, culturally legitimating and furthering the fortunes of autonomous and independent candidates, divorced from their parties and taking their appeals directly to voters (see Hersh, 2017).

This paper is part of 'A Manchurian candidate or just a dark horse? Towards the next generation of political micro-targeting research’, a Special issue of the Internet Policy Review.

Introduction

Political campaigns in the United States have employed digital technologies for more than a decade, developing increasingly sophisticated tools and techniques during each election cycle, as “computational politics” have become standard operating procedure (Tufecki, 2014; Kreiss, 2016). However, the most recent election marked a critical turning point, as candidates, political action committees, and other interest groups were able to take advantage of significant breakthroughs in data-driven marketing techniques, such as cross-device targeting, developed since the previous presidential election (“Bernie Sanders”, 2016; Edelman Digital, 2016). Electoral politics has now become fully integrated into a growing, global commercial digital media and marketing ecosystem that has already transformed how corporations market their products and influence consumers (Chahal, 2013; LiveRamp, 2015; Rubinstein, 2014; Schuster, 2015).The strategies, technologies, and tools of digital political marketing are more complex and far-reaching than anything we have seen before, with further innovations already underway (WARC, 2017). But because most commercial and political digital operations take place below the radar, they are not fully understood by the public. 1

In the following pages, we briefly describe the growth and maturity of digital marketing, highlighting its basic features, key players, and major practices. We then document how data-driven digital marketing has moved into the centre of American political operations, along with a growing infrastructure of specialised firms, services, technologies and software systems. We identify the prevailing digital strategies, tactics, and techniques of today’s political operations, explaining how they were employed during the most recent US election cycle. Finally, we explore the implications of their use for democratic discourse and governance, discussing several recent policy developments aimed at increasing transparency and accountability in digital politics.

Our research for this paper draws from our extensive experience tracking the growth of digital marketing over the past two decades in the United States and abroad, monitoring and analysing key technological developments, major trends, practices and players, and assessing the impact of these systems in areas such as health, financial services, retail, and youth (Chester, 2007; Montgomery, 2007; Montgomery, Chester, & Kopp, 2017). During the 2016 US presidential election, we monitored commercial digital advertising and data use by candidates, parties and special interest groups across the political spectrum. We collected examples of these ads, along with technical and market impact information from the developers of the applications. We also reviewed trade journals, research reports, and other industry documents, and attended conferences that were focused on digital technologies and politics. In the process, we identified all of the major providers of political digital data targeting applications (e.g., Google, Facebook, data clouds, ad agencies) and analysed all their key materials and case studies related to their 2016 operations. The source for much of this work was our ongoing gathering and analysis of cross-sectional commercial digital marketing practices worldwide.

Marriage of politics and commerce

Since the mid-20^th century, advertising has been an increasingly powerful and pervasive presence in US political campaigns, as a growing cadre of ad agencies, public relations firms, and consultants perfected the use of opinion polls, focus groups, and psychographics to reach and influence voters through radio, television, direct mail, and other media outlets (A. Jamieson, 2016; K. H. Jamieson, 1996; Sabato, 1981). With the rise of the internet, campaign operatives began to harness digital technologies and tools to mobilize voter turnout, engage young people, raise money, and support grassroots ground operations (Karpf, 2016; Kreiss, 2016; Tufecki, 2014). Both major political parties in the United States developed large, sophisticated data and digital operations (Kreiss, 2016).

Many of the digital strategies, tools, and techniques employed in the 2016 election were initially developed, deployed, tested, and refined by the commercial sector (Tufecki, 2014).Since its origins in the mid-1990s, digital marketing has operated with a core business model that relies on continuous data collection and monitoring of individual online behaviour patterns (Montgomery, 2011). This system emerged in the United States amid a political culture of minimal government interference, and within a prevailing laissez-faire ethos regarding the internet and new technologies (Barlow, 1996). In the earliest days of the “dot-com boom”, a strong political alliance was forged between the digital media companies and their partners in the advertising and media business, enabling the nascent industry to effectively ward off any attempts to restrain its business operations through privacy regulation or other public policies (Solon & Siddiqui, 2017). As a consequence, the advertising industry played a central role in shaping the operations of platforms and applications in the digital media ecosystem. Digital marketing is now well established and thriving, with expenditures reaching nearly $72.5bn in 2016 for the US alone, and worldwide spending predicted to reach more than $223bn this year (eMarketer, 2017; IAB, n.d.-d).

Ongoing innovations over the years have increased the capacity of data and digital marketing applications. Data collection, analysis, and targeting were further woven into the daily lives of consumers with the rise of social media platforms and mobile devices. Because of the unique role that they play in users’ lives, these platforms are able to sweep up enormous amounts of information, including not only what users post about themselves, but also what is collected from them throughout their daily activities (Smith, 2014). A growing arsenal of software and analytic tools has enhanced the ability of digital media companies and their advertisers to glean valuable insights from the oceans of data they generate (Smith, 2014). Predictive analytics introduced an expanded set of tools for scoring, rating, and categorising individuals, based on an increasingly granular set of behavioural, demographic, and psychographic data (“What is Predictive Intelligence”, 2017). US digital marketers have helped popularise and spur the successful adoption of digital advertising platforms and applications in nearly every geographical location with an internet connection or a link to a mobile device (IAB, n.d.-c). Google, Facebook, and other major players in the digital marketing industry have also developed a global research infrastructure to allow them, and especially their major advertising clients, to make continuous improvements in reaching and influencing the public, and to measure with increasing accuracy the success of their efforts (Facebook IQ, n.d.-a). These developments have created what some observers have called the “surveillance economy” (Singer, 2012).

The growth of data-driven political marketing

Though political campaigns have employed micro-targeting techniques—which use an array of personalised and other data sets and marketing applications to influence the actions of individuals—during the last several election cycles, recent technological innovations and industry advances have created a much more robust system than what was in place in 2012 (IAB, n.d.-b; Rubinstein, 2014). For years, political campaigns have been able to combine public voter files with commercial information from data brokers, to develop detailed and comprehensive dossiers on American voters (Rubinstein, 2014). With recent advances in the advertising technology and data industries, they can now take advantage of a growing infrastructure of specialty firms offering more extensive resources for data mining and targeting voters. Among the new entities are data marketing clouds. Developed by well-known companies such as Adobe, Oracle, Salesforce, Nielsen, and IBM, these clouds sell political data along with an exhaustive amount of detailed consumer information for each potential target, including, for example, credit card use, personal interests, consumption patterns, and TV viewing patterns (Salesforce DMP, 2017).

Some of these massive cloud services also operate what has become a new and essential component for contemporary digital targeting—the data management platform (DMP) (Chavez, 2017). DMPs provide marketers with “centralized control of all of their audience and campaign data” (BlueKai, 2011). They do this by collecting and analysing data about individuals from a wide variety of online and offline sources, including first-party data from a customer’s own record, such as the use of a supermarket loyalty card, or their activities captured on a website, mobile phone, or wearable device; second-party data, information collected about a person by another company, such as an online publisher, and sold to others; and third-party data drawn from thousands of sources, comprising demographic, financial, and other data-broker information, including race, ethnicity, and presence of children (O’Hara, 2016). All of this information can be matched to create highly granular “target audience segments” and to identify and “activate” individuals “across third party ad networks and exchanges”. DMPs are quickly becoming a critical tool for political campaigns (Bennett, 2016; Kaye, 2016, July; Regan, J., 2016).

Facebook and Google now play a central role in political operations, offering a full spectrum of commercial digital marketing tools and techniques, along with specialied ad “products” designed for political use (Bond, 2017). Not surprisingly, these companies have also made generating revenues from political campaigns an important “vertical” category within their ad business (Facebook, n.d.-d; Facebook IQ, n.d.-b; Stanford, 2016). Facebook’s role in the 2016 election was particularly important. With users required to give their real names when they sign up as members, Facebook has created a powerful “identity-based” targeting paradigm, enabling political campaigns to access its more than 162 million US users and to target them individually by age, gender, congressional district, and interests (Facebook, n.d.-b).Its online guide for political campaign marketing urges political campaigns to use all the social media platform tools it makes available to advertisers—including through Instagram and other properties—in order to track individuals, capture their data through various “lead-generation” tactics, and target them by uploading voter files and other data (Facebook, n.d.-a-c-f). The company also employs teams of internal staff aligned with each of the major political parties to provide technical assistance and other services to candidates and their campaigns (Chester, 2017; Kreiss & Mcgregor, 2017). Google heavily promoted the use of YouTube, as well as its other digital marketing assets, during the 2016 US election, reaching out to both major political parties (YouTube, 2017).

The growth and increasing sophistication of the digital marketplace has enhanced the capacities of political campaigns to identify, reach, and interact with individual voters. Below we identify seven key techniques that are emblematic of this new digital political marketing system, providing brief illustrations of how they were employed during the 2016 election.

Cross-device targeting

Getting a complete picture of a person’s persistent “identity” through an “identity-graph” has become a key strategy for successfully reaching consumers across their “omnichannel” experience (use of mobile, TV, streaming devices, etc.) (Winterberry Group, 2016). “Cross-device recognition” allows marketers to determine if the same person who is on a social network is also using a personal computer and later watching video on a mobile phone. Through data “onboarding,” a customer record that may contain a physical and email address is linked through various matching processes, associating it with what is believed to be that individual’s online identification—cookies, IP addresses, and other persistent identifiers (Levine, 2016). Cross-device targeting is now a standard procedure for political initiatives and other campaigns. Voter files are uploaded into the onboarding process, enabling the campaigns to find their targets on mobile devices and at specific times when they may be more receptive to a message (Kaye, 2016, April; L2, n.d.-b). Such granularity of information also enables a more tailored advertisement—so-called “dynamic creative”—which can be changed over time to “deliver very specific messaging” to individuals (Schuster, 2015). Leading cross-device marketing company Drawbridge offered a suite of election services in 2016 that provided campaigns a number of ways to impact voters, including through “Voter-Centric Cross Device Storytelling”, “Political Influencer Identification”, and via “Real-Time Voter Attribution Measurement” (Drawbridge, n.d.).

Programmatic advertising

Programmatic advertising refers to new automated forms of ad buying and placement on digital media using computer programmes and algorithmic processes to find and target a customer wherever she goes. The process can also involve real-time “auctions” that occur in milliseconds in order to “show an ad to a specific customer, in a specific context” (Allen, 2016). The use of programmatic advertising was one of the major changes in political campaign digital operations between 2012 and 2016—“the first time in American History”, according to one ad company, “that such precise targeting has ever been made available at such great scale” (Briscoe, 2017; Kaye, 2015). Programmatic advertising has itself grown in its capabilities to reach individuals, taking advantage of new sources of data to reach them on all of their devices (Regan, T., 2016). In 2016, for example, global ad giant WPP’s Xaxis system—“the world’s largest programmatic and technology platform”—launched “Xaxis Politics”. Capable of “reaching US voters across all digital channels”. the system is said to “segment audiences by hundreds of hot button issues as well as by party affiliation”, including via “real-time campaigns tied to specific real-world events” (Xaxis, 2015). Candidates were able to use the services of a growing list of companies, including Google, Rubicon, AOL, PubMatic, Appnexus and Criteo, that offered programmatic advertising platforms (“Political Campaigns”, 2016; Yatrakis, 2016).

Lookalike modelling

The use of big data analytics enables marketers to acquire information about an individual without directly observing behaviour or obtaining consent. They do this by “cloning” their “most valuable customers”in order to identify and target other prospective individuals for marketing purposes (LiveRamp, 2015). For example, Stirista (n.d.), a digital marketing firm that also serves the political world, offers lookalike modelling to identify people who are potential supporters and voters. The company claims it has matched 155 million voters to their “email addresses, online cookies, and social handles”, as well as “culture, religion, interests, political positions and hundreds of other data points to create rich, detailed voter profiles”. Facebook offers a range of lookalike modelling tools through its “Lookalike Audiences” ad platform. For example, Brad Parscale, the Trump campaign’s digital director, used the Lookalike Audiences ad tool to “expand” the number of people the campaign could target (Green & Issenberg, 2016). Facebook’s “Custom Audiences” product, similarly, enables marketers to upload their own data files so they can be matched and then targeted to Facebook users (Facebook, n.d.-e).

Geolocation targeting

Mobile devices continually send signals that enable advertisers (and others) to take advantage of an individual’s location—through the phone’s GPS (global positioning system), Wi-Fi, and Bluetooth communications. All of this can be done with increasing speed and efficiency. Through a host of new location-targeting technologies, consumers can now be identified and targeted wherever they go, while driving a car, pulling into a mall, or shopping in a store (Son, Kim, & Shmatikov, 2016). A complex and growing infrastructure of geolocation-based data-marketing services has emerged, with specialised mobile data firms, machine-learning technologies, measurement companies, and new technical standards to facilitate on-the-go targeting (Warrington, 2015). The use of mobile geo-targeting techniques played a central role in the 2016 election cycle, with a growing number of specialists offering their services to campaign operatives. For example, L2 (n.d.-a) made its voter file, along with HaystaqDNA modelling data, available for mobile device targeting, offering granular profile data on voters based on their interest in such contested topics as gun laws, gay marriage, voter fraud, and school choice, among others. Conde Nast’s Advance Publications’ election advertising worked with campaigns to append geo-location, profile data, and buying behaviour “to sculpt a very specific voter profile and target down to few hundred readers in a given geo location” (Ellwanger, 2016).

Online video advertising

Digital video, via mobile and other devices, is perceived as a highly effective way of delivering emotional content on behalf of brands and marketing campaigns (IAB, n.d.-a). There are a variety of online video ad formats that provide both short- and long-form content, and that work well for political and other marketing efforts. Progressive political campaign company Revolution Messaging, which worked for the Sanders campaign, developed what it calls “smart cookies” that it says take video and other ad placement “to the next level, delivering precision and accuracy” (Revolution Messaging, n.d.). Google’s YouTube has become a key platform for political ads, with the company claiming that today, voters make their political decisions not in “living rooms” in front of a television but in what it calls “micromoments” as people watch mobile video (DoubleClick, n.d.). According to the company’s political ad services research, mobile devices were used in nearly 60 percent of election-related searches during 2016. Content producers (which it calls “Creators”) on YouTube were able to seize on these election micro-moments to influence the political opinions of potential voters 18-49 (“Letter from the Guest Editors,” 2016).

Targeted TV advertising

Television advertising, which remains a linchpin of political campaign strategy, is undergoing a major transformation, as digital technologies and “addressable” set-top boxes have changed cable and broadcast TV into powerful micro-targeting machines capable of delivering the same kinds of granular, personalised advertising messages to individual voters that have become the hallmark of online marketing. Political campaigns are in the forefront of using set-top box “second-to-second viewing data”, amplified with other data sources, such as “demographic and cross-platform data from a multitude of sources” via information brokers, to deliver more precise ads (Fourthwall Media, n.d.; Leahey, 2016; NCC Media, n.d.). NCC Media, the US cable TV ad platform owned by Comcast, Cox, and Spectrum, provided campaigns the ability to target potential voters via the integration of its set-top box viewing information with voter and other data from Experian and others (Miller, 2017). Deals between TV data viewing companies and organisations representing both Republican- and Democratic-leaning groups brought the “targeting capabilities of online advertising to TV ad buys…bringing what was once accessible only to large state-wide or national campaigns to smaller, down-ballot candidates”, explained Advertising Age (Delgado, 2016).

Psychographic, neuromarketing, and emotion-based targeting

Psychographics, mood measurement, and emotional testing have been used by advertisers for many decades, and have also been a core strategy in political campaign advertising (Key, 1974; Packard, 2007; Schiller, 1975). The digital advertising industry has developed these tools even further, taking advantage of advances in neuroscience, cognitive computing, data analytics, behavioural tracking, and other recent developments (Crupi, 2015). Granular-based messages that trigger a range of emotional and subconscious responses, to better “engage” with individuals and deepen relationships with commercial brands, have become part of the DNA of digital advertising (McEleny, 2016). Facebook (2015), Nielsen, and most leading brands use “neuromarketing” services worldwide, which utilise neuroscience tools to determine the emotional impact of advertising messages. There is a growing field, recently promoted by Google, of “Emotion Analytics” that takes advantage of “new types of data and new tracking methods” to help advertisers “understand the impact of campaigns—and their individual assets—on an emotional level…” (Kelshaw, 2017). Scholars have identified that the use of “psychological targeting” in advertising enables the influencing of large groups of people by “tailoring persuasive appeals to the psychological needs” of the specific audiences (Matz, et al, 2017). Data company Experian Marketing Services for political campaigns offered data that weaved together “demographic, psychographic and attitudinal attributes” to target voters digitally. Experian claims its data enables campaigns to examine a target’s “heart and mind” via attributes related to their “political persona” as well as “attitudes, expectations, behaviours, lifestyles, purchase habits and media preferences (Experian, 2011, 2015). One of the most well publicised and controversial players in the 2016 election was Cambridge Analytica (CA), a prominent data analytics and behavioural communications firm that claimed to be a key component in Donald Trump’s victorious campaign. The company used a “five-factor personality model” aimed at determining “the personality of every single adult in the United States of America” (Albright, 2016; Kranish, 2016).Known as OCEAN, the model rated individuals based on five key traits: openness, conscientiousness, extroversion, agreeableness, and neuroticism. Drawing from digital data, voter history, and marketing resources supplied by leading companies, including Acxiom, Experian, Nielsen, GOP firm Data Trust, Aristotle, L2, Infogroup, and Facebook, CA was able to develop an “internal database with thousands of data points per person”. Its research also identified key segments that were considered “persuadable”, and shaped the advertising content placed “across multiple digital channels (with the most effective ads also appearing on television) (Advertising Research Foundation, 2017; Nix, 2016). The strategy was based on developing messages that were tailored to the vulnerabilities of individual voters (Nix, 2016; Schwartz, 2017). CA has become the subject of much scrutiny and debate, and itself has made conflicting claims, with critics raising concerns over its techniques and expressing scepticism about the extent of its impact (Confessore & Hakim, 2017; Karpf, 2017). However, the company’s work was sufficiently convincing to the leading advertising industry research organisation, the Advertising Research Foundation (2017, March), that it honoured the firm with a “Gold” award in 2017 under its “Big Data” category.

Discussion

The above description provides only a brief overview of the data-driven marketing system that is already widely in use by candidate and issue campaigns in the United States. The increasingly central role of commercial digital marketing in contemporary political campaigns is reshaping modern-day politics in fundamental ways, altering relationships among candidates, parties, voters, and the media. We acknowledge that digital technologies have made important positive contributions to the vibrancy of the political sphere, including greatly expanding sources of news and information, significantly increasing opportunities for citizen participation, and empowering people from diverse backgrounds to form coalitions and influence policy. The same tools developed for digital marketing have also helped political campaigns substantially improve voter engagement, enhance their capacities for “small-donor” fundraising, and more efficiently generate turnout (Moonshadow Mobile, n.d.; Owen, 2017). However, many of the techniques we address in this paper raise serious concerns—over privacy, discrimination, manipulation, and lack of transparency.

Several recent controversies over the 2016 election have triggered greater public scrutiny over some of the practices that have become standard operating procedure in the digital media and marketing ecosystem. For example, “fake news” has a direct relationship to programmatic advertising, the automated system of “intelligent” buying and selling of individuals and groups (Weissbrot, 2016). These impersonal algorithmic machines are focused primarily on finding and targeting individual consumers wherever they are, often with little regard for the content where the ads may appear (Maheshwari & Isaac, 2016). As a consequence, in the middle of the 2016 election, many companies found themselves with ads placed on “sites featuring pornography, pirated content, fake news, videos supporting terrorists, or outlets whose traffic is artificially generated by computer programs”, noted the Wall Street Journal (Nicas, 2016; Vranica, 2017). As a major US publisher explained in the trade publication Advertising Age,

Programmatic’s golden promise was allowing advertisers to efficiently buy targeted, quality, ad placements at the best price, and publishers to sell available space to the highest bidders…. What was supposed to be a tech-driven quality guarantee became, in some instances, a “race to the bottom” to make as much money as possible across a complex daisy chain of partners. With billions of impressions bought and sold every month, it is impossible to keep track of where ads appear, so “fake news” sites proliferated. Shady publishers can put up new sites every day, so even if an exchange or bidding platform identifies one site as suspect, another can spring up (Clark, 2017).

Criticism from news organisations and civil society groups, along with a major backlash by leading global advertisers, led to several initiatives to place safeguards on these practices (McDermott, 2017; Minsker, 2017). For example, in an effort to ensure “brand safety”, leading global advertisers and trade associations demanded changes in how Google, Facebook and others conduct their data and advertising technology operations. As a consequence, new measures have been introduced to enable companies to more closely monitor and control where their ads are placed (Association of National Advertisers, 2017; Benes, 2017; IPA, 2017; Johnson, 2017; Liyakasa, 2017; Marshall, 2017; Timmers, 2015).

The Trump campaign relied heavily on Facebook’s digital marketing system to identify specific voters who were not supporters of Trump in the first place, and to target them with psychographic messaging designed to discourage them from voting (Green & Issenberg, 2016). Campaign operatives openly referred to such efforts as “voter suppression” aimed at three targeted groups: “idealistic white liberals, young women and African Americans”. The operations used standard Facebook advertising tools, including “custom audiences” and so-called “dark posts”—“nonpublic paid posts shown only to the Facebook users that Trump chose” with personalised negative messages (Green & Issenberg, 2016). Such tactics also took advantage of commonplace digital practices that target individual consumers based on factors such as race, ethnicity, and socio-economic status (Google, 2017; Martinez, 2016; Nielsen, 2016). Civil rights groups have had some success in getting companies to change their practices. However, for the most part, the digital marketing industry has not been held sufficiently accountable for its use of race and ethnicity in data-marketing products, and there is a need for much broader, industry-wide policies.

Conclusion

Contemporary digital marketing practices have raised serious issues about consumer privacy over the years (Schwartz & Solove, 2011; Solove & Hartzog, 2014). When applied to the political arena, where political information about individuals is only one of thousands of highly sensitive data points collected and analysed by the modern machinery of data analytics and targeting, the risks are even greater. Yet, in the United States, very little has been done in terms of public policy to provide any significant protections. In contrast to the European Union, where privacy is encoded in law as a fundamental right, privacy regulation in the US is much weaker (Bennett, 1997; Solove & Hartzog, 2014; U.S. Senate Committee on Commerce, Science, and Transportation, 2013). The US is one of the only developed countries without a general privacy law. As a consequence, except in specific areas, such as children’s privacy, consumers in the US enjoy no significant data protection in the commercial marketplace. In the political arena, there is even less protection for US citizens. As legal scholar Ira S. Rubinstein (2014) explains, “the collection, use and transfer of voter data face almost no regulation”. The First Amendment plays a crucial role in this regard, allowing the use of political data as a protected form of speech (Persily, 2016).

The political fallout over the how Russian operatives used Facebook, Twitter, and other sites in the 2016 presidential campaign has triggered unprecedented focus on the data and marketing operations of these and other powerful digital media companies. Lawmakers, civil society, and many in the press are calling for new laws and regulations to ensure transparency and accountability for online political ads (“McCain, Klobuchar & Warner Introduce Legislation”, 2017). The U.S. Federal Election Commission, which regulates political advertising, has asked for public comments on whether it should develop new disclosure rules for online ads (Glaser, 2017). In an effort to head-off regulation, both Facebook and Twitter have announced their own internal policy initiatives designed to provide the public with more information, including what organisations or individuals paid for political ads and who the intended targets were. These companies have also promised to establish archives for political advertising, which would be accessible to the public (Falck, 2017; Goldman, 2017; Koltun, 2017). The US online advertising industry trade association is urging Congress not to legislate in this area, but to allow the industry to develop new self-regulatory regimes in order to police itself (IAB, 2017). However, relying on self-regulation is not likely to address the problems raised by these practices and may, in fact, compound them. Industry self-regulatory guidelines are typically written in ways that do not challenge many of the prevailing (and problematic) business practices employed by their own members. Nor do they provide meaningful or effective accountability mechanisms (Center for Digital Democracy, 2013; Gellman & Dixon, 2011; Hoofnagle, 2005). It remains to be seen what the outcome of the current policy debate over digital politics will be, and whether any meaningful safeguards emerge from it.

While any regulation of political speech must meet the legal challenges posed by the First Amendment, limiting how the mining of commercial data can be used in the first place can serve as a critically important new electoral safeguard. Advocacy groups should call for consumer privacy legislation in the US that would place limits on what data can be gathered by the commercial online advertising industry, and how that information can be used. Americans currently have no way to decide for themselves (such as via an opt-in) whether data collected on their finances, health, geo-location, as well as race or ethnicity can be used for digital ad profiling. Certain online advertising practices, such as the use of psychographics and lookalike modelling, also call for rules to ensure they are used fairly.

Without effective interventions, the campaign strategies and practices we have documented in this paper will become increasingly sophisticated in coming elections, most likely with little oversight, transparency, or public accountability. The digital media and marketing industry will continue its research and development efforts, with an intense focus on harnessing the capabilities of new technologies, such as artificial intelligence, virtual reality, and cognitive computing, for advertising purposes. Advertising agencies are already applying some of these advances to the political field (Facebook, 2016; Google, n.d.-a; Havas Cognitive, n.d.). Academic scholars and civil society organisations will need to keep a close watch on all these developments, in order to understand fully how these digital practices operate as a system, and how they are influencing the political process. Only through effective public policies and enforceable best practices can we ensure that digital technology enhances democratic institutions, without undermining their fundamental goals.

Disclaimer: This guest essay in the Special issue on political micro-targeting has not been peer reviewed. It is an abbreviated version of a speech delivered by the Member of the European Partiament (MEP) Sophie in ‘t Veld in Amsterdam in May 2017 to Data & Democracy, a conference on political micro-targeting.

Democracy

Democracy is valuable and vulnerable, which is reason enough to remain alert for new developments that can undermine her. In recent months, we have seen enough examples of the growing impact of personal data in campaigns and elections. It is important and urgent for us to publicly debate this development. It is easy to see why we should take action against extremist propaganda of hatemongers aiming to recruit young people for violent acts. But we euphemistically speak of 'fake news' when lies, 'half-truths’, conspiracy theories, and sedition creepily poison public opinion.

The literal meaning of democracy is 'the power of the people'. 'Power' presupposes freedom. Freedom to choose and to decide. Freedom from coercion and pressure. Freedom from manipulation. 'Power' also presupposes knowledge. Knowledge of all facts, aspects, and options. And knowing how to balance them against each other. When freedom and knowledge are restricted, there can be no power.

In a democracy, every individual choice influences society as a whole. Therefore, the common interest is served with everyone's ability to make their choices in complete freedom, and with complete knowledge.

The interests of parties and political candidates who compete for citizen’s votes may differ from that higher interest. They want citizens to see their political advertising, and only theirs, not that of their competitors. Not only do parties and candidates compete for the voter's favour. They contend for his exclusive time and attention as well.

Political targeting

No laws dictate what kind of information a voter should rely on to be able to make the right consideration. For lamb chops, toothpaste, mortgages or cars, for example, it’s mandatory for producers to mention the origin and properties. This enables consumers to make a responsible decision. Providing false information is illegal. All ingredients, properties, and risks have to be mentioned on the label.

Political communication, however, is protected by freedom of speech. Political parties are allowed to use all kinds of sales tricks.

And, of course, campaigns do their utmost and continuously test the limits of the socially acceptable.

Nothing new, so far. There is no holding back in getting the voters to cast their vote on your party or your candidate. From temptation with attractive promises, to outright bribery. From applying pressure to straightforward intimidation.

Important therein is how and where you can reach the voter. In the old days it was easy: Catholics were told on Sundays in church that they had no other choice in the voting booth than the catholic choice. And no righteous Catholic dared to think about voting differently. At home, the father told the mother how to vote. The children received their political preference from home and from school. Catholics learned about current affairs via a catholic newspaper, and through the catholic radio broadcaster. In the Dutch society, which consisted of a few of such pillars, one was only offered the opinions of one's own pillar1. A kind of filter bubble avant la lettre.

Political micro-targeting

Nowadays, political parties have a different approach. With new technologies, the sky is the limit.

Increasingly advanced techniques allow the mapping of voter preferences, activities, and connections. Using endless amounts of personal data, any individual on earth can be reconstructed in detail. Not only can their personal beliefs be distilled from large troves of data, no, it even is possible to predict a person's beliefs, even before they have formed them themselves. And, subsequently, it is possible to subtly steer those beliefs, while leaving the person thinking they made their decision all by themselves.

As often is the case, the Americans lead in the use of new techniques. While we Europeans, touchingly old-fashioned knock on doors and hand out flyers at Saturday's market, the American employ the latest technology to identify, approach, and influence voters.

Of course, trying to find out where voters can be reached and how they can be influenced is no novelty. Political parties map which neighbourhoods predominantly vote for them, which neighbourhoods have potential, and in which neighbourhoods campaigning would be a wasted effort. Parties work with detailed profiles and target audiences, for which they can tailor their messages.

But the usage of personal data on a large scale has a lot more to offer. Obviously, this is a big opportunity for political parties, and for anyone else, who runs campaigns or aims to influence the elections.

However, the influencing techniques become increasingly opaque. As a result of the alleged filter bubble, voters are being reaffirmed in their own beliefs, and they hardly receive information anymore about the beliefs and arguments of other groups. This new kind of segmentation may stifle critical thinking. There may not be enough incentive to test one's own ideas, to find new arguments, or to critically reflect on the truthfulness of information.

I am a social and economic liberal D66 politician, and I get suggestions for news articles from websites like The Guardian or Le Monde. My colleague from the right wing nationalist PVV, may well receive URLs from Breitbart.

Pluralism is essential for a healthy, robust democracy. In a polarised society, people live in tightly knit groups, which hardly communicate with each other. In a pluralist society people engage in the free exchange, confrontation, and fusion of ideas.

The concept pluralism is under pressure. Populist parties declare themselves representative of The People. In their vision, The People, is uniform and homogenous. There is a dominant cultural norm, dictated from the top-down, to which everyone must conform. Whomever refuses, gets chewed out. Often, it is about one-dimensional symbolism such as Easter eggs and Christmas trees. There is no place for pluralism in the world of the populists. But when there is no pluralism, there is no democracy. Without pluralism, democracy is nothing more than a simple tribal dispute, instead of the expression of the will of all citizens together.

Voter data

European privacy legislation limits the use of personal data. In the world of ‘big data’, one of the explicit goals of regulation is to prevent restriction of the consumer's choice. Oddly enough, lawmakers do not explicitly aspire to guarantee voters as broad a choice as possible. But in politics, individual choices have consequences for society as a whole.

In 2018, the General Data Protection Regulation (GDPR) comes into effect. We have worked five years on the GDPR. At this moment, we work on the modernisation of the e-Privacy Directive, which is mainly about the protection of communication. As was the case with the GDPR, companies from certain sectors scream bloody murder. European privacy protection would mean certain death for the European industry. According to some corporate Cassandras, entire European industries will move to other continents. That very same death of corporate Europe is also predicted for any measure concerning, say, environmental norms, procurement rules, or employee rights. All those measures are in place, but, as far as I know, the nightmare scenario has never occurred...

There are some corporate sectors, such as publishing and marketing, which have a huge impact on the information supply to citizens. They are the ones who now cry wolf. It is understandable that they are unhappy with stricter rules concerning their activities, but as the potential impact of the use of personal data and ‘big data’ increases, so does their social responsibility.

At the moment, there is not much public debate about the new techniques. Peculiar. Thirty years ago, 'subliminal advertising', as we called it then, was prohibited because people found it unethical to influence people without their knowledge. We need to have a similar debate. What do we think of opaque influencing? Do we need ethical norms? Should such norms apply only to political campaigns, or should we look at this from a broader perspective? In the ‘big data’ debate, we tend to speak in technical or legal terms, while actually the issue is fundamentally ethical, holding far-reaching consequences for the vitality of our democracy.

Such a public debate demands more clarity on the impact of ‘big data’, profiling, targeting, and similar techniques on the individual, her behaviour, and her choices, which determine in what direction society progresses. Which voters are being reached? How sensitive are they for the subtle influencing and what makes them resilient? How do people who are hardly reached only compare to the others? How do voters and non-voters compare? Is the voter truly predictable? Can we identify or influence the floating voter? Do voters actually float between different parties? Or do they especially float within their own party, their own bubble, their own segment? How important are other factors, such as the social context? If the new influencing techniques are indeed as potent as we think, how can polls get it so wrong? What can we learn from advertisers who return to contextual advertising, because targeting turns out less effective than they thought?

We need to stay cool-headed. New technologies have a huge impact, but human nature will not suddenly change due to ‘big data’ and its use. Our natural instincts and reflexes will definitely not evolve in a few years. That would take many thousands of years, as even in the 21st century, we seem to have more than a few cavemen traits, so losing internalised behaviour is not as easy as 1-2-3. Humans are resilient, but democracy is vulnerable. On a short term, the societal impact is large. This gives us all the reason to reflect on how to deal with the new reality, and how we can keep up our values in this new reality.

The use of personal data, clearly, is not solely reserved for decent political parties. Other persons and organisations, from the Kremlin to Breitbart, can bombard European voters with information and misinformation. But European governments, controlling endless amounts of personal data of their citizens, can also manipulate information, or circulate utter nonsense to advance their own interests. A random example: the Hungarian government influencing their voters with lies and manipulation about the so-called consultation on asylum seekers.

Beyond voter data

This issue is not only about the personal data of voters, but also about the personal data of political competitors, opponents, and critics, which are increasingly being employed. Recently, we have seen efforts of external parties to influence the results of the 2017 French elections. We saw a large-scale hack of the Emmanuel Macron campaign, and the spread of false information, coming obviously from the Kremlin and the American Alt-Right, meant to discredit Macron's candidacy.

Also, the American elections show the shady game of hacking, leaking, and manipulating. The issue of the Hillary Clinton mails will undoubtedly occupy our minds for years. Who knows how the elections would have turned out without this affair?

Other democratic pillars can get corrupted as well by the misuse of data. Critical voices, opposition, and checks and balances are democracy's oxygen. Democracy is in acute jeopardy when data are employed to attack, undermine, discredit, blackmail, or persecute journalists, judges, lawyers, NGOs, whistleblowers, and opposition parties.

In Europe, we tend to shrug our shoulders at these dangers. "Oh well, we'll see, such things occur only in banana republics, not right here". Of course, this trust in our democratic rule of law is wonderful. But if we treat our rule of law this neglectfully, we will lose it eventually.

Within the European Union, we currently see this happening in Poland and Hungary. The governments of both nations ruthlessly attack independent judges, critical media, inconvenient NGOs. They do so with quasi-lawful means. Under the banner of transparency, they force NGOs to register. In doing so, they misuse laws against money laundering, and terror finance. Or the governments bring out compromising information about judges or politicians in strategic moments.

But critical voices struggle in other member states as well. Lawyers are being monitored, even without a legal basis. In the years after 9/11, we have created endless new abilities for intelligence services, police and justice departments to spy on citizens, even without suspicion, without the signature of a judge. The companies to which we unwittingly surrender our personal data, in exchange for service, are forced to hand over all information to the government, or forced to build in backdoors. Governments hack computers in other countries. Usually, it starts out with unlawful practices, but soon enough laws are put in place to legalise those practices. The magic word 'terrorism' silences any critique on such legislation.

But when politicians, journalists, NGOs, whistleblowers, lawyers, and many others cannot perform their tasks freely and without worry, our democracy withers. Not only do they have to operate without someone keeping an eye on them, they have to know nobody is in fact watching them. The mere possibility of being watched, results in a chilling effect.

For this principal reason, I have contested a French mass surveillance law before the French Conseil d'Etat. Since, as a member of the European Parliament, I spend four days a month on French soil (in Strasbourg), I could potentially be the target of the French eavesdropping programme. This is not totally imaginary, as I am not only a politician, but also a vocal critic of certain French anti-terror measures. It is not about me actually worrying about being spied on, but about the fact that I might be spied on. Luckily, I am not easily startled, but I can imagine that many politicians are vulnerable. That is a risk for democracy.

I do not discard the possibility of a ruling of the European Court of Human Rights on my case. In that turn of events, it will lead to jurisprudence valid in the entire EU (and the geographical area covered by the Council of Europe).

But, of course, this should not depend on the actions of one obstinate individual whether politicians, NGOs, journalists, and so on, can do their jobs fearlessly, to fulfil their watchdog role.

It is my personal, deep, conviction that the biggest threat to our democracy is the fact that we have enabled the powerful to access, with almost no limitations, the personal data of those who should control those very same powerful entities.

What can we do?

Some propose new forms of democracy, in which universal suffrage is weakened or even abolished. In his book ‘Against elections: the case for democracy’, David Van Reybrouck had the idea to appoint representatives on the basis of chance, and in his book ‘Against democracy’ Jason Brennan wants to give the elite more votes than the lower classes, presuming that people with more education or development make better choices. Others want to replace representative democracy with direct democracy.

I oppose those ideas. Universal suffrage and the representative democracy are great achievements, which have led to enormous progress in society.

First of all, we have to make sure our children grow up to be critical, independent thinkers. Think differently, deviate, provoke: this must be encouraged instead of condemned. A democracy needs non-conformists.

We must teach our children to contextualise information and to compare sources.

The counterpart of ‘big data’ must be ‘big transparency’. We need to understand not just open administration, but also insights into the techniques of influence.

The regulation and limitation of the use of personal data, as I hope to have argued effectively, is not a game of out-of-touch privacy activists. It is essential for democracy. We need safeguards, not only to be sure people really are free in their choices, but also to protect the necessary checks and balances. As such, I plea for a rigorous application of the GDPR, and in the European Parliament, I will work for a firm e-Privacy Directive.

And yes, perhaps we should examine whether the rules for political campaigning are still up-to-date. In most countries, those rules cover a cap on campaign expenditures, a prohibition of campaigning or polling on the day before election day, or a ban on publishing information that may influence the election results, such as the leaked e-mails in France. But these rules have little impact on the use of personal data to subtly influence elections.

Last year, the European Parliament supported my proposal for a mechanism to guard democracy, the rule of law, and fundamental rights in Europe.2

On this day (editor’s note: 9 May, Europe Day) of European democracy, I plead for equal, high norms in Europe. The last years have shown that national elections are European elections. It is crucial for us to trust that all elections in EU member states are open, free, and honest elections, free of improper influencing.

These last sixty years, the European Union has developed itself into a world leader in democracy and freedom. If we start a public debate, Europe can remain a world leader.

This paper is part of 'A Manchurian candidate or just a dark horse? Towards the next generation of political micro-targeting research’, a Special issue of the Internet Policy Review.

Introduction

As political campaigns compete, they try to outsmart each other by all sorts of actions: from dropping witty puns during a televised debate, to strategically knocking on doors and convincing voters. Technological innovation can help political parties improve the effectiveness of their campaigns. By using technology to collect, process, and analyse information about voters, campaigns can improve their knowledge about the electorate. Subsequently, technology can extend campaigns' capabilities of targeting specific groups with tailored messages resulting in more efficient campaigning. We call this phenomenon 'political behavioural targeting' (PBT).

Several scholars have researched political behavioural targeting in the US context (e.g., Kreiss, 2012, 2016; Nielsen, 2012; Hersh, 2015). However, the US differs in several obvious ways from most European countries. One can imagine that differences in electoral systems, privacy laws, and party financing influence campaigns’ ability to collect, process, and use personal voter data. Therefore, the findings from these studies do not necessarily apply to European countries. As there is little research in a European context, it remains unclear to what extent and how campaigns in a multiparty democracy, such as the Netherlands, use PBT-techniques. Also, it is unclear if and why there are differences between parties. In line with Colin Bennett (2016, p. 261), we wonder: "can political parties campaign in Europe as they do in North America?"

Such a question is relevant, as some scholars fear that the use of data and targeting techniques hinders public deliberation (Gorton, 2016), weakens the mandate of elected officials (Barocas, 2012), has negative effects on citizens’ privacy (Howard, 2006; Rubinstein, 2014; Tene, 2011), and enables campaigns to send tailored messages directly to citizens, thereby avoiding scrutiny from journalists (Jamieson, 2013). As a result, campaigns can potentially make opposite promises to different people, without anyone noticing.

This article sheds light on how Dutch political campaigns adopt and use PBT-techniques. Through interviews with campaign leaders, using a grounded theory approach, we answer the following overarching research question: What barriers and facilitators for the adoption and use of PBT-techniques do Dutch political parties perceive?

Theoretical framework

We will first summarise innovations in political campaigns over time, leading up to the advent of political behavioural targeting. Then, we identify the factors influencing the adoption of PBT on a campaign team level. Finally, we explore the factors that can shape the adoption of PBT on the level of national systems.

Innovations in political campaigns

Political campaigns have continuously been adapting to technological developments. Pippa Norris (2000) describes how the advent of television and the shift from partisan newspapers to national television news triggered a process of modernisation in the way political campaigns operated. Notable consequences of this shift were the adoption of a media-centred strategy in order to set the agenda, the rise of political marketing, the collection and use of data (such as opinion polls) to "shape, fine-tune and monitor campaign efforts" (Blumler, Kavanagh, and Nossiter, 1996; p. 53). Another shift came with the internet and the new possibilities for party-voter interaction that came along with the medium, which led campaigns to a new stage of the modernisation process: the postmodern campaign (Norris, 2000).

It would be an oversimplification to point to 'the internet' as a game-changer in political communication, because of the rapidly changing nature of the internet itself. As David Karpf (2012, p. 640) notes: "the internet of 2002 has important differences from the internet of 2005, or 2009, or 2012". Accordingly, much more than the advent of the internet itself, it is the advent of social media such as Facebook (2004), YouTube (2005) and Twitter (2006) which provided political campaigns with new ways of communication with the electorate (e.g., Gibson & McAllister, 2011; Conway, Kenski, & Wang, 2015; Vaccari, 2012). Together with companies such as Google, whose core business is actually not its well-known search engine but rather its advertisement business, social media not only facilitate new ways of communication, but also the tracking and collection of behavioural data of internet users (Zuiderveen Borgesius, 2016). This technique ("behavioural targeting") originates from the advertisement business. Ad agencies monitor people's online behaviour and combine this information with consumer data provided by data brokers, to target them individually with tailored ads (Turow, 2011, p.75). When applying this concept to the political realm, we can dub this phenomenon as political behavioural targeting (PBT).

Of course, PBT is not about selling products but about winning votes. And political campaigns have different means to do so than advertisement agencies have (e.g. canvassing efforts); which means that PBT happens offline as well as online. We distinguish traditional canvassing from PBT-canvassing if campaigns are able to process information about individual conversations (such as the voter's likelihood to vote for a party or her most important voting consideration), and subsequently use that information to gain strategic insights about the electorate and/or to target the voter at a later stage with a tailored message, while skipping the 'wrong' doors in a neighbourhood (Kreiss, 2016; Nielsen, 2012).

Arguably, the use of PBT can be seen as the latest step within the modernisation of political campaigns. However, as we have seen in earlier phases, not all parties in all countries adopt new techniques at the same pace and rate. Below, we identify the factors influencing the adoption of PBT. We organise these factors at two levels: (1) the individual campaign around a candidate/party and (2) the national system (i.e., the electoral system, regulatory framework, and culture). This translates into the model shown in Figure 1, which will be elaborated on in the next paragraphs.

Figure 1: Factors influencing the adoption of PBT

The campaign team level

In his extensive research of US political campaigns, Daniel Kreiss (2016) identified four factors concerned with technological innovation within political campaigns. There are resource factors, such as campaign budgets and the number of volunteers a campaign can employ; infrastructural factors, such as technological tools or skills within the organization; organizational factors, such as organisational culture and structure; structural electoral cycle factors, such as election results. Building upon Kreiss' factors, we add an additional four (one campaign team level factor and three system level factors) to examine the use of PBT. On a campaign team level, the factor is ethical and legal concerns, such as normative reservations towards PBT. On a system level, the factors are electoral context, regulatory framework,and culture (discussed below). These new factors were identified through a review of literature about innovation in data-driven political campaigning techniques (e.g., Anstead, 2017; Kreiss, 2016; Jungherr, 2016; Hersh, 2015; Nielsen, 2012), and literature about (hybridisation of) campaign evolvement (e.g., Lijphart, 2012; Plasser & Plasser, 2002; Karlsen, 2010; Norris, 2000).

Resource factors

The main elements within this factor that could influence the extent to which campaigns can use PBT-techniques are: the budget and the effort needed to carry out a PBT-operation. A large budget enables campaigns to hire skilled personnel, acquire data, or buy targeted ads. The same dynamic applies to the number of volunteers a campaign can mobilise: having a lot of them facilitates a campaign in collecting data by canvassing, and sending potential voters targeted messages (the use of volunteers, of course, is dependent on their skills). Having a small budget and few volunteers, consequently, can be a barrier for campaigns because it bars them from acquiring the same amount of capabilities or from carrying out an operation on a large scale. This is in line with normalization theory (Margolis & Resnick, 2000), according to which the possibilities of the internet will not upset traditional power structures, but will rather develop along traditional lines as in the 'offline world'.

We can also view PBT as a means of using a campaign's resources as efficient as possible, to ensure parties do not spend money and effort on voters who will vote for another party anyway, or on citizens who will not vote altogether. Then, parties with limited resources could be more inclined to use PBT to not waste precious money, time, and labour. This is in accordance with the idea of equalization, which views the internet as an empowering tool for smaller parties due to its low costs and its new ways of direct communication with the electorate (Margolis, Resnick, & Levy, 2003; Bimber & Davis, 2003; Stanyer, 2010). A meta-analysis found evidence for the existence of both normalisation and equalisation in election campaigns (Strandberg, 2008). The occurrence of either process can differ per country and is dependent on several contextual factors, which will be discussed later on.

Organisational factors

The elements in this factor are about how campaign leaders perceive campaigning. Do they rely on proven best practices from previous campaigns or is there a culture of innovation? John Padgett and Walter Powell (2012) describe the concept of network folding. Applied to the political realm, this entails the extent to which campaigns employ skilled personnel from non-political sectors and to integrate that expertise into their existing institutions. An example is the hiring of Google engineer Stephanie Hannon as chief technology officer by the Clinton campaign (Easton, 2015). The 'cognitive diversity' following from network folding can lead to creative ideas (De Vaan, Stark, & Vedres, 2015). Furthermore, the organisational structure can be expected to resemble the way the campaign perceives PBT. A campaign with an autonomous data department is probably more prone to rolling out a PBT-strategy than a campaign that sees 'data' as only one of the many tasks of a communication staffer. Also, a change in leadership can be a facilitator for innovation (Gibson & Römmele, 2001).

Infrastructural factors

Elements are the technological tools available to campaigns, which enable them to roll out a PBT-operation. For instance, such tools might assist volunteers in the field by enabling them to collect data. They can be developed in-house or outsourced; in fact, there are specialised third party consultancies, that offer off-the-shelf tools, which in turn allow campaigns to employ innovative technology even though the campaigners do not have any technical expertise.

Structural electoral factors1

The actions of rival campaigns fall under the umbrella of structural electoral factors. A successful PBT-campaign of a rival can facilitate innovation in other campaigns, especially if those other campaigns themselves look back at an unsuccessful election. This connects with the 'critical event' (Kreiss, 2016), such as losing an election that should have been won, or with the experience of an 'external shock', which can be an incentive for professionalisation (Gibson & Römmele, 2001).

A second element influencing campaigns' likelihood to use PBT-techniques, is issue ownership (Petrocik, 1996), and the subsequent statements of party candidates propagating standpoints of the party. A political campaign 'caught' using privacy-infringing PBT-techniques, while its candidates present themselves as privacy champions, is likely to come across as hypocritical. Being perceived as such should be avoided, considering the negative electoral consequences following political-ideological hypocrisy perpetrated by politicians (Bhatti, Hansen, & Olsen, 2013).

Ethical and legal concerns

Elements within this factor consist of ethical and legal restrictions on how campaigns operate. For example, a political party could believe that PBT is ethically wrong as it infringes on citizens' right to privacy, and citizens' autonomy to form their own opinions. As a result, the party 'self-regulates' and refrains from using campaigning techniques violating its ethical beliefs.

Another element is the legal uncertainty that occurs when a campaign does not know how to behave in accordance with data protection and election laws, because of a lack of internal expertise. Such confusion can result in differences in the actions taken by comparable actors (e.g. Raskolnikov, 2017). Legal uncertainty can lead to 'overcompliance', which can be seen as a barrier towards the adoption of PBT-techniques, or to 'undercompliance', which facilitates the adoption of PBT-techniques (Calfee & Craswell, 1984). For instance, Anstead (2017) notes how parties felt disadvantaged by targeting possibilities facilitated by the perceived undercompliance with UK campaign finance law during the 2015 general elections.

The system level

Aside from campaign level factors, we look at contextual factors as well. These factors may limit the extent to which (US-American) campaigning techniques can be adopted in other countries (Karlsen, 2010). Therefore, we add three new contextual factors to our model. We expect that the electoral system, the regulatory system, and the culture of a democracy influence the extent to which the campaign team level factors are applicable. Below, we explore how the adoption of PBT-techniques can be influenced by properties of different systems. We will later apply our model (see Figure 1) to one specific case.

Electoral system

The three dominant electoral systems are first-past-the-post (FPTP), proportional representation (PR), and two-round (TR) (Birch, 2001, 2003). How these systems function, can influence how campaigns are run. The FPTP-system, first, can lead to an overvaluation of some key districts. Such districts sometimes 'swing' to one party and sometimes to another party, whereas other districts go to the same party in each election. As an effect, campaigns in a FPTP-system are inclined to spend a disproportionate amount of money and labour in these key districts in the hope of swinging the election their way (e.g., Anstead, 2017; Lipsitz, 2004). The PR-system, second, does not favour a select group of voters in a few key districts (Plasser & Plasser, 2002). This is especially true when the PR-system consists of only one district, in which every vote counts equally. As a result, campaigns have to spread their means more equally over the country. The TR-system, third, makes for a relatively unpredictable campaign, since it often is unclear which candidates will make it to the second round. Furthermore, the TR-system makes it important for campaigns to collect the votes of the supporters of the losing candidates of the first round. Therefore, campaigns should not only focus on their own base but other candidates' bases as well (Blais & Indridason, 2003). This has consequences for PBT, since campaigns should not only correctly classify potential voters as their own, but the other voters as well in order to target them in the next round.

A different aspect of electoral systems that influences how a campaign is run, is the degree of fractionalisation in a democracy (Duverger, 1959; Lijphart, 2012; Wang, 2012). FPTP-systems favour relatively few candidates/parties. PR-systems, in contrast, enable a large number of parties to run in an election. The first round of a TR-system can consist of many different candidates. As a result, campaigns that operate in a PR or a TR-system are less likely to launch attack campaigns against competitors. This is because PR-systems generally require a coalition of parties working together after the elections (Plasser & Plasser, 2002). And in a TR-system, campaigns should not mistreat competing candidates too much because winning campaigns have to court the bases of losing candidates in order to win the second round. Furthermore, in a highly fractionalised democracy, parties represent different (minority) groups within the electorate. This results in a high risk of 'mistargeting', in which campaigns approach a member of group A with appeals for a member of group B. Mistargeting can lead to voters penalising the campaign for their mistake (Hersh, 2013). These contextual circumstances may call for different PBT-strategies.

Regulatory framework

We distinguish strictly regulated campaign environments, moderately regulated campaign environments and minimally regulated campaign environments (Plasser & Plasser, 2002). Strictly regulated environments are characterised by "severe restrictions on the contact and communication between candidates and their constituencies" (p. 137). Moderately regulated environments typically focus on regulating access to TV-advertising and campaign funding. Minimally regulated environments impose few regulatory restrictions on political campaigns. It may be infeasible to implement innovative PBT-techniques in strictly regulated environments. Legal uncertainty can play a role on a system level too (e.g. because of a gap in the law).

Culture

Differences in the adoption and use of innovative PBT-practices can also be influenced by the culture or tradition in a democracy. For example, turnout culture is important because campaigns operating in countries where turnout is high will focus more on convincing voters on getting out the vote than campaigns operating in a low-turnout culture. In a low-turnout culture, campaigns sometimes target specific groups of voters (e.g. the elderly, who are more likely to turn out) more than other groups (e.g. the poor), who are unlikely to turn out (Herrnson, 2001). The turnout culture can influence the data campaigns collect on someone (and how campaigns tailor their messages), because a campaign message meant to convince someone typically leans on more data than a message meant to mobilise a voter does. Furthermore, cultural norms can dictate the strategy of political campaigns. In Japan, for example, posting dark post attack ads, such as the 'super predator' ad Trump launched against Clinton (Green & Issenberg, 2016), is improbable because of the cultural convention of averting direct conflict (Plasser & Plasser, 2002).

System level context is likely to affect campaign level factors. A campaign operating in a multiparty PR-system needs to pour more resources into identifying potential supporters than a campaign in a FPTP-system. After all, identifying potential Republicans or Democrats is easier than identifying potential voters in a ten-party race. Moreover, other than in US campaigns, most European campaigns are unable to access voter registration files provided by an electoral register. In many countries citizens can just show up at the voting booth, which means that the whole act of ‘registering’ for voting, as it is the case in the US, does not exist. Since Hersh (2015) has found that voter lists are one of the most valuable pieces of data to US campaigns, this principal unavailability, or non-existence, of such data poses a challenge for the PBT-capabilities of campaigns. But this challenge should by no means imply a PBT-operation in Europe is impossible. We would argue that while the lack of access to voter lists makes it very difficult to achieve the same level of granularity when engaging in PBT as in the US, by using other commercially available or self-collected data, it can be possible to come reasonably close to the desired level (which may be more modest for European campaign leaders than for American ones). The extent to which there is an actual difference between the degree of granularity between US and European campaigns, however, is outside of the scope of this paper, as we focus on perceptions and strategies of campaign leaders.

System level context also affects infrastructure. For instance, should the groundwork be spread equally across the nation, or focused on a number of battleground states? Also, a campaign operating in a heavily regulated context is likely to encounter some legal barriers. For example, because of campaign financing regulations (may influence resource factors), and data protection regulations (infrastructural factors). The absence of regulations, conversely, can facilitate PBT. Cultural context, finally, can influence campaigns' ethical considerations regarding PBT. Campaigns operating in a culture that favours privacy, for example, can be expected to avoid (or use less-invasive) PBT-techniques than campaigns run in a culture in which privacy is less important. In sum, there are several factors, both on campaign and system level, which can form a barrier or facilitate the extent to which campaigns are able to use PBT-techniques and how they use them.

Extending existing research to a European context, we have developed and will apply an improved model (applicable in different electoral contexts) to analyse barriers and facilitators to innovative PBT-practices by political campaigns. As the context of the research case differs from the US, we expect to contribute to the framework and to shed light on how contextual factors influence innovation of political campaigns. Furthermore, in answering our research question, we provide insight into the way political campaigns in a multiparty democracy organise, communicate and innovate. Given these considerations, our key question is: What barriers and facilitators for the adoption and use of PBT-techniques do Dutch political parties perceive?

Method

This study focuses on campaigns in the Netherlands because of the national elections taking place in the research period (15 March 2017), the advanced technological infrastructure (Coy, 2015), and the interesting contextual factors. The Dutch electoral system is one of open list proportional representation (PR), in which all members of parliament come from one nationwide district (Lijphart, 2012). This means that in the Netherlands, every vote counts equally. Moreover, the system of PR (and the very low de-facto threshold) enables a relatively large number of political parties to run in an election. 28 parties participated in the 2017 national election ("Partijen nemen deel", 2017). Of these parties, 13 actually gained a seat in parliament ("Officiële uitslag", 2017).

The Dutch national elections have a relatively high turnout: around 80% in the previous two elections ("Officiële uitslag", 2017). But where the US presidential campaigns can spend hundreds of millions of dollars (Narayanswamy, Cameron, & Gold, 2017), the Dutch campaign with the biggest funds (VVD) has no more than 5 million dollars to spend. And even if the budgets were sufficiently large, the question is whether voter data would be usable for a political campaign. The Dutch data protection law categorises political preference as sensitive personal data. This means that campaigns are only allowed to process such information if the potential voter explicitly gives permission to do so.

Finally, as party membership in the Netherlands steadily decreases, political campaigns can rely less on their members to do labour-intensive tasks (such as canvassing). In 2016, the number of party members of all political parties combined, was at its lowest point since the second world war. Although this number has picked up slightly since, party membership is still quite low ("Membership Dutch parties still low", 2017).

After approval from the ethical committee of the University of Amsterdam, we carried out eight in-depth interviews with campaign leaders. We interviewed 11 campaign leaders, belonging to eight political parties in total (three interviews were double-interviews). In addition to this, we held two background interviews (with one local campaigner for the municipality of Amsterdam and one political consultant offering PBT-services). The eight elite-interviews on average lasted 53 minutes. Two were conducted by phone, the others face-to-face. We took a qualitative research approach for several reasons: the small group of people concerned with the coordination of political campaigns in the Netherlands, the lack of knowledge on this topic in the Netherlands, and because interviewing is a suitable method for understanding the mechanisms behind and perceptions of a phenomenon (Boeije, 2005). As we want to understand how campaigns see PBT, what they are actually doing, and how they perceive possible barriers and facilitators to the adoption and use of PBT-techniques, the interview is a suitable data-collection method. Using an interview guide (see appendix A), we held semi-structured interviews, allowing follow-up questions.

Interviewees

We selected the interviewees via purposive sampling. Campaign leaders qualified for an interview when they had a coordinating role in the campaign and were campaigning for a party that gained at least one seat in the 2012 national parliamentary elections. Eleven campaigns satisfied this second criterion (see Table 1). We contacted interviewees via email, explaining the objective of the study. The interviewees signed an informed consent document before the interview started. We also promised the campaign leaders anonymity, and confidentiality until after election day (15 March 2017). By doing so, we tried to provide the interviewees with a safe environment in which they felt like speaking freely, and without concern of somehow 'leaking' strategic information. Because no information would become public before election day, the risk that interviewees might provide biased information due to a strategic agenda was minimised. Another large advantage of interviewing the campaign leaders before the elections took place, is the prevention of hindsight bias by the interviewees. Unfortunately, we were unable to convince three parties to comply (VVD [right on the political spectrum], PVV [right-wing nationalist party], PvdD [Party for the animals; left-wing]). These three parties were unwilling to cooperate, either because they still found the risk of leaking their strategy too large, or they did not offer an explanation.

Analysis

Using a grounded theory approach, this study has passed four phases: the exploration phase, the specification phase, the reduction phase, and the integration phase (Wester, 1995). In the exploration phase, two background interviews took place (with a campaigner for the municipality of Amsterdam and with a political consultant offering PBT-services). These were coded using AtlasTI, 'tentatively labelling' relevant information (Glaser, 1978). Thereafter, the first interviews with campaign leaders took place. These were transcribed and open-coded. Furthermore, fellow researchers also coded these interviews and discussed the content (peer debriefing). In the next phase, new interviews took place and the data was subject to axial coding. The first dimensions were identified (e.g., what forms a barrier and what facilitates the use of PBT-techniques?). The reduction phase saw the emergence of the core category (innovation). In the integration phase, we completed the conceptual framework, finalised our analysis, and had the campaign leaders approve the quotes used (member checking). This means the campaign leaders agreed with the way they were quoted, and with the publishing of the names of the political parties. Member checking increased the willingness of campaign leaders to cooperate with the study.

Results

We first describe the field: to what extent do campaigns use PBT-techniques? Then we explain differences between parties by focusing on the five campaign level factors concerning the use of PBT (resource, infrastructural, organisational, structural electoral cycle factors, and ethical and legal concerns). Finally, we zoom out to the system level and discuss the influence of contextual factors (electoral system, regulatory framework and culture) on the adoption of PBT-techniques.

PBT in Dutch campaigns

As campaigns in the Netherlands can have recourse to relatively detailed public census data and detailed election results, all campaigns, to some extent, adopt a PBT-approach. Furthermore, Facebook is an important tool for all parties, but the parties differ in how they use Facebook's capabilities. Some parties occasionally post content targeted to broad age groups, while other campaigns frequently post content tailored to more specific groups. Two campaigns stand out, as they have developed their own PBT-tools, which they can use to continuously refine their knowledge of the electorate. We will now use our model to explain the differences between campaigns.

Resource factors

All campaigns cite financial costs as a barrier. Table 2 shows that budgets are modest, and differ between parties.

Source: parties' annual financial reports, on file with authors.

These small budgets form a barrier for the cooperation with expert political consultants (such as Blue State Digital) to enhance their PBT-operations. Parties refer to the financial costs as the main reason not to hire consultants. Campaign leader 1 of the Social Democratic Party (PvdA) explains why he does not work with Blue State Digital (BSD):

Their system is very expensive, that's a factor. And you need the people to carry out the work for you. In an ideal world, such a cooperation would be really cool though."

Liberal democrat party D66 agrees: "because it costs a lot of money and we don't have that kind of money. And if we spend it on a consultant, we can't spend it on the campaign itself."

The same barrier appears when campaigns speak about other technological means, such as canvassing apps, allowing campaigns to directly process information from canvassers. Christian Democrats CDA, for instance, would like such an app. Campaign leader 8: "yes, but that would demand a financial investment that we can't afford."Green Party GroenLinks has a contrasting perspective: "I believe it usually costs around €100,000 to build an app such as our own. (..) We, however, paid our programmers two crates of beer and 40 pizzas." Several facilitators help GroenLinks and also socialist party SP to overcome this barrier of financial costs. First: the personal network of the campaign leader. This facilitator is especially prominent for GroenLinks, where campaign leader 7 employs his own network to optimise the BSD-systems, but also to help him with setting up other parts of the campaign:

We had to adjust it [the BSD system] somewhat for the Netherlands. The people with whom I did so, Swedish folks ... they are simply a little network of people of around my own age, and some people who are a bit older and have already set up a similar campaign in their own countries. A guy who set up the grassroots organization for Trudeau, for example, he's a couple of years older than I am, but I Skype with him to talk about how I should handle certain things.

Campaign leader 7's personal network plays (or at least played)an important role in cheaply setting up technological tools and creating content:

Through the network, I'm aware of the crowdfunding streams for a normal campaign. (..) I'm meeting a friend tomorrow, who has experience with mail flows. (..) I have a network of volunteering writers, poets, freelance journalists who write for us for free. (..) So partly, I just have a good personal network.

However, as PvdA notes, having lots of data is of no use if you don't have the capacity to use it. GroenLinks tries to overcome this barrier by organising their campaigns, to some extent, in a citizen-initiated manner (Gibson, 2015). A citizen-initiated campaign (CIC), devolves "power over core tasks to the grassroots" (p.183). As campaign leader 7 puts it:

Grassroots is about creating an infrastructure to enable as many sympathizers as possible to volunteer as canvassers on a large scale. So voter contact on a large scale, but also – and that's Bernie's [Sanders] lesson – to have places in which a few people make stuff by themselves without us having any control over it. (..) Embracing people's creativity without managing it.

SP has less need of a citizen initiated campaign, because of their relatively large number of active party members. "The big difference [with GroenLinks] is the fact that we already have the volunteers. Many other parties lack the numbers. We have thousands of party members who gladly canvass for us two weekdays and on Saturday as well." [Campaign leader 11, SP].

Infrastructural factors

Having a good infrastructure allows campaigns to actually collect data and send tailored messages. What kind of PBT-infrastructure can parties rely on and how does it facilitate their use of PBT?

All campaigns use the PBT-infrastructure Facebook offers, although some more than others. Nearly all campaigns use its lookalike audiences function to find new potential voters. Campaign leader 8: "we search for profiles of people who look like the ones who’ve already liked our Facebook page, and then serve them with advertisements." Campaigns also look at people who like pages that are close to the values of the political parties. Christian party ChristenUnie, for instance, tries to target voters who like the page of evangelical broadcaster EO. So does the Calvinist Political Party SGP, which tries to find out people’s interests on Facebook: "For example… farming, or Israel, off the top of my head; you try to approach people along the lines of their interest, or the region in which they reside." [Campaign leader 9]

Some campaigns also employ ‘dark posts’, a Facebook function that enables campaigns to opaquely target specific audiences, while its messages are not visible to untargeted Facebook users. Campaign leader 1 exemplifies:

We’ve managed to get something done related to gas extraction in Groningen. It doesn’t make sense to share that on the national Facebook page, because it was only important news locally. So we put out a dark post, only for Groningen residents. Sometimes we can specify it even more.”

Using Facebook for PBT-purposes, campaigns do not actually gather or own data themselves. There are a few campaigns that do gather their own data, by using canvassing apps. Campaign leader 7:

We use the election results per voting location and use that information to establish the GroenLinks mindedness of a neighbourhood. Then we can prioritise which addresses to visit and which to ignore. When we visit addresses, our volunteers use the app to answer the following questions: 1. Is anyone home? 2. Does she want to talk? 3. Is she going to vote? 4. Is she planning to vote for GroenLinks? 5. What is the most important theme to her? 6. How GroenLinks minded was she? If she considers to vote for GroenLinks, two questions follow: 1. Do you want to stay informed of our campaign by e-mail? 2. Can I have your phone number, so we can ask you to do canvassing talks?”

The GroenLinks app facilitates large scale collection of information about people's political preferences, thereby informing strategic decisions. Also, the personal data can facilitate accurate PBT on an individual level. The secondary objective of the app is to provide an infrastructure for volunteers to campaign on their own terms, whenever they feel like doing so:

Our app, built by hackers, enables others to campaign for us. (..) Someone in [small town] Lutjebroek can install our app and go ahead and work for our campaign. No campaign leader needed. [Campaign leader 7]

Some campaigns monitor the visitors of their own websites. Campaign leader 1: “What are people searching for on our website, how do they get to our website, how much time do they spend, (..) which button should you colour red? How does that work?” At the time of the interview, CDA was not yet tracking their website visitors, but: “we’ve just migrated to a new website, on which we want to start collecting more data on our visitors. I’m curious what kinds of people are visiting the website. And what kinds of people don’t, and therefore have to be reached through different channels.”

SP has built a system which combines previous election results, census data and their own membership Constituent Relationship Management (CRM) data. Plotted on a Google Map, they can identify interesting areas for them to canvass. This system facilitates efficient use of means:

We would do nothing more happily than knocking on every single door in every city, but unfortunately, we do not yet have that kind of manpower. So we do an analysis: What kinds of neighbourhoods are especially interesting for us? We have built our own system to help us make that decision [Campaign leader 11].

Organisational factors

Circumstances within the campaign’s organisation itself can form a barrier for the uptake of PBT-techniques. Less innovative parties, for instance, do not have a dedicated data, tech, or digital department. As a campaign leader notes: "The department responsible for that [tech/data/digital] is our Communication department. So that's four or five people. And sometimes someone of the department picks it up, but there's not one specific person who's responsible."This contrasts with GroenLinks, which has a Digital and Grassroots department and with SP's Digital department.

The "state of mind"within a campaign can also be seen as a barrier:"In the sense that internally, people are still very much inclined to think offline. The culture within the campaign is quite offline." [anonymous campaign leader]2

New leadership and younger staffers can play a facilitating role in political organisations. Campaign leader 11 argues that, because he is young, their new party chair brings a more tech-savvy vision than his predecessor. According to the campaign leader, younger staffers are more likely to implement tech and data in their work procedures.

A final organisational barrier is the primary goal a political party pursues. Campaign leader 10:

Maybe the strange thing about SGP is that we do not care that much about seat maximization. For us, it's about the impact of our principles. And sure, we would rather have four seats than three, but if we have to settle for three seats: that's fine too. And that's, in my opinion, a reason why we have a feeling like: do we really need data?

Structural electoral cycle factors

These circumstances are largely beyond the control of the campaigns, but they can influence the uptake of PBT-techniques. Campaign leaders see the PBT-actions of other political campaigns as a motivational factor. As campaign leader 11 notes about the development of their app: "I've looked a little bit at how GroenLinks have their app and canvassing system." Or as campaign leader 2 concludes: "If every party does it, you don't win very much by it. But if you're the only party that does nothing..."

Ethical and legal concerns

Especially D66 and the seniors’ party 50PLUS take a principled stance against the collection of data and the use of PBT. Where D66 presents itself as a privacy champion and therefore will never gather and use information about (groups of) voters, 50PLUS campaign leader 6 warns about the risk of irresponsible use of the data gathered by the "almost stalking of people", which he calls "morally irresponsible".

Furthermore, a lack of internal legal expertise appears to contribute to a feeling of legal uncertainty, which affects the likelihood of adopting PBT-techniques: "Legislation has grown so very comprehensive and complex. It's almost impossible to cope for us as a small organisation." [Campaign leader 5]

While ethical and legal concerns can form a barrier, a left- or right-wing orientation does not seem to be instrumental therein. After all, we have seen left-wing parties GroenLinks and SP develop relatively advanced PBT-tools. And we have seen right-of-centre party CDA express clear interest in advancing their own PBT capabilities. At the same time, left-of-centre 50PLUS and right-of-centre D66 both oppose the use of PBT.

System level

Electoral system

Although the Dutch one district PR-system should make for a rather equal distribution of campaign efforts, campaigns still divide the country into smaller areas of interest called 'key areas'. These areas differ per party, but do receive a relatively large part of campaign attention. Campaign leader 1 describes these as areas: "where we know the turnout is low, but the number of PvdA-voters is high". All campaigns use data provided by the Electoral Council, showing the election results per party, per voting location to establish key areas. Campaign leader 3 explains:

Using that [the election results], you see: Okay, we do well in this neighbourhood or this street. And then you combine that information with the CBS3 data, to find out what kind of neighbourhood it is, what kind of people live there, what are their backgrounds, how much do they earn, what does the family composition look like, et cetera.

Facilitated by these public data, campaigns enrich their knowledge of specific areas. A next step would be to use those data to make personalised appeals to (subgroups of) people living in those specific key areas.

Regulatory framework

Although the Netherlands would qualify as a minimally regulated environment (Plasser & Plasser, 2002; Esser & Strömbäck, 2012), campaigns all experience regulatory pressure and legal uncertainty on a system level. They cite an abundance of regulations, forming a barrier to their ability to innovate.

The technological developments have been taking place so very quickly. And, in that timeframe, to adjust all your procedures and everything. And also to meet the privacy regulations, I think many parties face a huge challenge in that respect." [Campaign leader 4]

Campaigns sometimes face a dilemma, having to decide between innovative techniques and privacy regulations. Campaign leader 11:

Regulations sometimes are unclear, which leads us to decide to go for the safe option because you do not know where the red line is. And you never want to abuse someone's personal data. So yes, regulations sometimes cause us to hit the brake and that's a good thing.

Culture

There is a recurring worry about the perceived low level of political knowledge of the average Dutch voter. PBT-techniques can facilitate campaigns' efforts to convince or educate such low-information voters, for example by "having a conversation with someone, especially if you share some characteristics," [Campaign leader 7] or by interesting "people for things that are relevant to them and to make them aware of the political dimension of those things." [Campaign leader 8] Campaign leader 3, in contrast, concludes that the electorate's low level of political knowledge (together with the perceived volatility of the electorate, the decrease in political trust, and their focus on persons instead of parties) forms an insurmountable barrier, making PBT-techniques irrelevant.

Discussion and conclusion

In the 2017 elections, used here as a case study, all campaigns use PBT through Facebook, but some parties are more advanced than others, and have even developed their own PBT-tools. We have established what the main barriers and facilitators for PBT are, using five factors on a campaign level and three factors on a system level. Not only does this study shed light on the conditions under which these barriers and facilitators manifest themselves, it also gives insight into their different workings across parties. Our study provides information about the data collected by parties and the PBT-techniques used to attract voters. We demonstrate how personal networks and cognitive diversity within a campaign can level barriers. We show how PBT is not only perceived as useful for campaigns in a FPTP-system, but in a PR-system as well. And we show how regulatory pressure is perceived as an obstacle and as a welcome 'normative red line'.

A triangulated research approach can improve our understanding of the campaign leaders' constructs. Observation of their (use of) PBT-tools and how these tools help campaigns make strategic decisions, can give more insight into the workings of these techniques. Another approach would be to interview canvassers and identify 'field-level' barriers and facilitators. Furthermore, ideally, we would have spoken to all parties holding a seat in parliament. Unfortunately, three parties did not cooperate. Two of those became the largest (VVD) and second largest (PVV; in a very close field) party. Since we did have access to eight of 11 parties, we are confident about our findings and we do not expect to identify additional factors influencing the adoption of PBT from interviews with the remaining parties.

Compared to related recent studies by Anstead (2017), Hersh (2015), Kreiss (2016), and Nielsen (2012), this study makes a number of contributions. In general, we focus our exploratory research on a PR-system instead of a FPTP-system, and we develop a model that takes system level contextual factors into account. Specifically, unlike Anstead (2017) we have found evidence for equalisation (which occurs when smaller parties take advantage of the internet's low costs and direct communication possibilities, and, in doing so, use the internet as a tool of empowerment [e.g. Margolis, Resnick, & Levy, 2003]). This evidence is especially clear in the case of GroenLinks, which was, at the time of the campaign, one of the smallest parties in parliament (now the fifth party). Furthermore, we provide an insightful point of view into Anstead's question of whether "parties develop data-driven capabilities more rapidly in electoral systems with a tendency towards disproportionate outcomes" (2017; p. 23). In comparison with Hersh (2015), we focus less on how differences in data-availability lead to different strategic decisions, and more on how differences in the perception of campaign level and system level factors lead to a variation in the occurrence in PBT-innovation. With regard to Kreiss (2016), we have extended his model and applied it to a multiparty democracy. In comparison with Nielsen (2012), we focus solely on the perception of campaign leaders and not on canvassers. Furthermore, we focus on PBT on online as well as offline platforms.

Our attention for the system level factors has enabled us to identify perceived influence of the PR-system on the adoption of PBT. Contrary to theoretical expectations (Plasser & Plasser, 2002), campaigns in a one-district PR-system do identify key-areas that are more heavily campaigned than other districts. These key-areas differ from 'battleground states' in FPTP-systems in the sense that the key-area does not sometimes swing one way and sometimes the other, but rather that potential voters in key-areas are supportive of a certain party, but not very likely to show up at the polls. Campaigns use PBT-techniques to convince these potential voters of the personal relevance of politics and to motivate them to cast their vote. Areas with firm turnout numbers and clear support for a certain party, in contrast, are perceived as less decisive and less of a priority. This leads to a hierarchy of areas, which differs per party. Also, as a PR-system typically leads to a relatively large number of parties partaking in an election, PBT can be seen as an asset for a campaign to organise in a more efficient manner. Moreover, according to the campaign leaders, PBT-techniques offered by Facebook do allow smaller parties a degree of visibility that they are unable to achieve through traditional media.

On a campaign level, in the coming years, we expect more citizen-initiated campaigning (Gibson, 2015) by campaigns low in labour-resources. This requires a solid infrastructure, which opens the door for third party intermediaries offering off-the-shelf infrastructure. In this regard, it would be interesting to track the development of PvdA, which has suffered its biggest loss in history. This critical event could lead to the prototyping (Kreiss, 2016) of GroenLinks' innovative campaign by PvdA. As the party's chairman has resigned, the door is open to a more cognitive diverse party structure (Du Pre, 2017; De Vaan et al., 2015). Of course, these developments might apply less to parties that are officially more cognisant of campaign ethics (e.g. D66). This is why ethics and legal aspects are important factors to take into consideration. It would be interesting to see how these campaigns act as PBT-capabilities of rival parties improve. Their self-imposed barrier can limit their future chances, but can also attract voters growing more aware of the value of privacy. In the former case, this could lead to an overhaul of their privacy principles, or perhaps to a legislative push towards the restriction of PBT (similar to Hersh, 2015). In the latter case, campaigns can be expected to develop innovative non privacy-invasive campaigning techniques. Either way, our model would provide tools to study the process.

So 'can political parties campaign in Europe as they do in North America' (Bennett, 2016)? We would say ‘mostly yes’. We agree with Bennett (2015) that there are important differences between the US and Europe, and indeed, they influence how PBT is used. But based on our findings, we are hesitant to conclude that those differences (severely) constrain the export of PBT-practices to European multiparty systems. We have shown that relatively small campaign budgets do not need to bar parties from engaging in PBT-practices (or even from cooperating with BSD, an 'expensive' American political consultancy). The same is true of the electoral system: campaign leaders generally perceive PBT-techniques as useful in a PR-system. What remains is the relatively strict Dutch data protection law, labelling political preference as 'sensitive personal data', which can only be processed with explicit consent from the potential voter. 'Explicit consent', however, sounds harsher on paper than it is in practice and is easily achieved (e.g., Beales & Muris, 2008; Calo, 2012; Joergensen, 2014). Of course, because of data regulations and/or their non-existence, European campaigns are unable to consult voting lists showing whether an individual showed up at the polls in the last elections. In most European countries, the electoral register is inaccessible to political parties. One might argue that, from a campaign's perspective, US voter data are superior to European voter data. We would argue that European data are different, but they do not bar European campaigns in the use of PBT-techniques. Dutch campaigns, for instance, can (and do) rely on election results on voting booth level (which comprises a couple of streets). They can (and do) combine these results with detailed, accurate, and a multitude of data about the neighbourhoods surrounding those voting booths. And then there is Facebook, facilitating easy targeting of its users with personalised messages. As potential challenges for democracy come with PBT, such as ignoring 'less valuable' citizens (e.g. reliable non-voters), more research into the workings and effects of PBT is needed.

Disclaimer: This guest essay in the Special issue on political micro-targeting has not been peer reviewed. It is treated here as a reflection.

During the past three decades there has been a persistent, and dark, narrative about political micro-targeting. Phil Howard (2006) vividly described a present and future where politicians would use data to “redline” the citizens that received political information, manufacturing attitudes and beliefs, leading to “managed citizenship”. In the years since Howard wrote his monumental book, the concerns over micro-targeting have only grown. The explosion of data about the electorate in Western democracies such as Australia, Canada, the UK, and the United States (Howard & Kreiss, 2010) has triggered deep unease among scholars and privacy advocates alike. Sophisticated voter databases now contain everything from political party data gleaned through millions of interactions with the electorate, public data obtained from state agencies, and commercial marketing information that is bought and sold on international open markets. The 2016 US presidential election revealed the new ways that individuals can be profiled, identified, found, tracked, and messaged to on social media platforms such as Facebook and YouTube, which these companies themselves help facilitate (Kreiss and McGregor, 2017).

While it might seem that the micro-targeting practices of campaigns have massive, and un-democratic, electoral effects, decades of work in political communication should give us pause. Although we lack the first-hand data from political campaigns, consultancies, and technology firms such as Facebook to know for sure, previous research tells us that people are seldom the unwitting dupes of strategic political communication. Partisanship shapes much of how people vote and decades of research reveals that it is really hard to change people’s minds through campaigns (Kalla & Broockman, 2017; Henderson & Theodoridis, 2017). This has large implications for the effectiveness of micro-targeting. For example, Eitan Hersh’s (2015) deeply and carefully researched ground-breaking study using data from a major vendor to the US Democratic Party discovers that campaign practitioners find it really hard to persuade voters. This is because political practitioners lack reliable and identifiable data on cross-pressured and low information voters. Given this, campaigns often focus on known voters rather than risk targeting and messaging to the wrong people. Indeed, Hersh reveals that despite hundreds of data points on members of the electorate, it is a small cluster of publically available data – such as turnout history, party identification, and demographic data – that matters far more for predicting vote choice.

The lesson is that micro-targeted campaign ads are likely most effective in the short run when campaigns use them to mobilise identified supporters or partisans, spurring volunteerism, donations, and ultimately turnout – hardly the image of a managed, manipulated, or duped public (Baldwin-Philippi, 2017). Ironically, campaigns often use micro-targeting to further these forms of democratic participation, making appeals to targeted subsets of voters on the basis of the parties and issues they already care about. Campaigns also use micro-targeting in the attempt to decrease voter turnout on the opposing side, sending negative messages to the oppositions’ likely voters in the hopes this will make them less excited to turn out for their candidate. But two decades of social science suggests that this can be a risky strategy given that partisans can rally behind their candidate who is being attacked (Dunn & Tedesco, 2017).

What explains the outsized concerns about micro-targeting in the face of the generally thin evidence of its widespread and pernicious effects? This essay argues that we have anxieties about micro-targeting because we have anxieties about democracy itself. Or, to put it differently, that scholars often hold up an idealised vision of democracy as the standard upon which to judge all political communication. In a world where many scholars and journalists both hope and ardently believe, in the face of all available evidence, that members of the public are fundamentally rational, seek to be informed, and consider the general interest, micro-targeting appears to be manipulative, perverting the capacity of citizens to reason about politics. Meanwhile, for many scholars and journalists, political elites are fundamentally opposed to members of the public, seeking domination or control as opposed to representing their interests. In this world, much of the concern over micro-targeting reads as a classic “third-person effect”, where scholars and journalists presume that members of the public are more affected by campaign advertising than they themselves are.

And yet, this idealised version is not how democracy really is, nor necessarily how it should be. The argument of this brief essay is that, as a quantifiable practice premised on strategically identifying targeted groups of voters and crafting messages designed to appeal to them, micro-targeting is broadly reflective of the fact that democracy is often partisan, identity-based, and agonistic – in short, political. Following communication scholar Michael Schudson’s (1986) study of commercial advertising nearly three decades ago, this essay asks the following questions in the US context: what is the work that micro-targeting does, where does it fit into the political culture, and, what kind of political culture has given rise to it? I argue that micro-targeting is only imaginable, and efficacious, in a polity that prizes partisan mobilisation, group solidarity, agonism, and the clash of opposing moral views in its politics. Following from this, I suggest different democratic concerns about micro-targeting that relate to its cultural power to, over time, create a powerful set of representations of democracy that undermines the legitimacy of political representation, pluralism, and political leadership.

The cultural work of micro-targeting

To analyse the role that micro-targeting plays in politics, first we need to understand how and why citizens vote. In their recent book Democracy for Realists, political scientists Christopher Achen and Larry Bartels (2016) offer a sustained critique of what they call the “folk theory” of American democracy. According to this “folk theory” that underlies conceptions of popular sovereignty, Americans have identifiable and consistent policy preferences. During the course of an election, they inform themselves about the policy positions of candidates and make rational decisions as to which best represents their preferences, which in turn leads parties to be responsive to the wishes of the public.

As Achen and Bartels (ibid.) argue, this is a fiction. They outline a “group theory of democracy”, where it is social attachments and group identification that largely determine both partisanship and vote choice. Achen and Bartels argue that people see themselves in relation to the groups that they belong to and those that they do not. Identity is so strong, in this account, that it conditions both what partisans believe parties stand for but also their interpretation of facts (ibid., 267; see also Prasad et al., 2009). As Achen and Bartels demonstrate, this identity and group theory of politics has expansive empirical support over seventy years of research which demonstrates, time and again, that people have little knowledge about politics and yet detailed understandings of the social groups that the Democratic and Republican parties are perceived to represent. It is in this context that candidate performances of partisan and social identity become more important for electoral outcomes then the informational content of journalism. Events and candidates make identity more or less salient and strengthen group attachments. During campaigns, parties and candidates work to remind voters of their partisan and social attachments and strengthen them so they are mobilised to participate in the election. As Achen and Bartels (ibid., 311) argue:

Political campaigns consist in large part of reminding voters of their partisan identities – “mobilizing” them to support their group at the polls. Formal communications by the groups and informal communication networks among group members also help citizens understand how their identity groups connect to the candidates and parties.

In this context, what is important about political campaigns is this work of communicating the partisan and social identities of candidates to voters. Candidates and their campaigns use micro-targeting, along with other strategic communications, to accomplish this. Micro-targeting is both a campaign practice of using data to craft and deliver strategic messages to subsets of the electorate (historically across many different media), and a genre of campaign communications that, much like political advertising more broadly, reinforces and amplifies the partisan, group, and identity conflicts at the heart of US politics. There has been extensive research on how micro-targeting works as a data-driven and quantifiable practice (see, for instance, Karpf, 2016). What these messages do as a genre of campaign communications, however, has received considerably less scrutiny. Drawing on my own previous work in the US context (Kreiss, 2016), the first argument that I develop here is that micro-targeting furthers the mobilisation that Achen and Bartels (2015) identify, primarily through reminding citizens of and shoring up their partisan and group identities. I then discuss the potential democratic consequences of this in a more expansive, cultural sense.

Micro-targeted ads have an aesthetic of what I call “political realism”, building on Michael Schudson’s work on commercial advertising. In Advertising, The Uneasy Persuasion, Schudson (1986) compared commercial advertising with Soviet realist art (the official state-sanctioned art of the former Soviet Union), arguing that it offers a form of “commercial realism”. As commercial realism, commercial advertising “simplifies and typifies” (215); advertising is abstracted, presenting the world as it should be, not as it is, and it exemplifies individuals as members of larger social groups. As it does so, “the aesthetic of capitalist realism — without a masterplan of purposes — glorifies the pleasures and freedoms of consumer choice in defense of the virtues of private life and material ambitions.” (ibid., 218) 

We can see micro-targeted digital advertising as a cultural form of ‘political realism’ that reflects, reinforces, and celebrates a political culture, at least in the United States, premised on identity, moral certainty, and mobilisation - not weighty considerations of the general interest or deliberation. Micro-targeted digital content shares a few central characteristics, which I adapt here for politics from Schudson’s (1986) work on commercial realism:

• It presents social and political life in simplified and typified ways;
• It presents life as it should become, or for negative ads, as it must not become;
• It presents reality in its larger social significance, not in its actual workings;
• It presents progress towards the future and positive social struggle, or for negative ads, the ideas of the other party as negative steps back into the past. It carries a message of optimism for one partisan side, and takes a stance of pessimism towards political opponents; and,
• It tells us that political conflict is necessary, a clash of different groups and worldviews; moral certainty is assured, political identity is certain, and political agonism is reality.

For example, micro-targeted ads present social life in simplified ways, not presenting actual lives but abstract, stylised ones designed to be rife with larger meaning. A depiction of a farmer’s daily work in a campaign ad, for instance, is not about actual events or daily labours, but is meant to be an abstract, simplified, symbol of the American values of hard work and cultivation of the earth and celebration of ordinary people in a democratic society. The farmer here is typified; the campaign ad is not about a real person who farms. The farmer is a representation of the larger social categories, values, and ideas the ad presents as desirable or worthy of emulation for all Americans. At the same time, the two dominant US political parties often stress different themes in their ads, a recognition that they have different visions of what life should be become, what progress is, and what worldviews and moral claims the public should embrace. While doing so, political micro-targeting is inherently pluralist. It reflects a basic claim that “everyone has interests to defend and opinions to advance about his or her own good, or the group’s good, or the public good, and every interest was at least potentially a political interest group.” (Rosenblum, 2010, 259)

While it is impossible to know the full range of micro-targeted ads run during the course of an election cycle, consider some of the examples culled from the non-profit and non-partisan Democracy in Action website that chronicles US campaigns and the Hillary for America Design 2016 website that compiles the creative design from the campaign. To start, much of political micro-targeting is about building campaign databases by finding supporters online, signing them up for the cause through email, and repeatedly messaging them to enlist them in becoming a volunteer or a donor.

Take, for instance, the declarative “I am a Hillary Voter” digital ad (see Figure 1), presumably (but also logically) directed (although we can never know for sure) at the candidate’s supporters. What separates micro-targeted political ads from their mass broadcast counterparts is the data that lies behind them: campaigns can explicitly try to find and send messages to their partisan audiences or intra-party supporters, linking the names in their databases to identities online or on social media platforms such as Facebook. Campaigns can also try to find additional partisans and supporters by starting with the online behaviours, lifestyles, or likes or dislikes of known audiences and then seeking out ‘look-alike audiences’, to use industry parlance. And, what people do when they see these ads is quantified in terms of their performance, measured through things such as engagement and click-throughs. Micro-targeting is about mobilisation through conveying and building social solidarity. While there is much concern over candidates speaking out of both sides of their mouths to the electorate through hyper-targeted digital ads, likely far more often campaigns use micro-targeting to provide occasions for social identification and group belonging, conveying and constructing the sense of shared identity and group membership at the heart of politics. The “Wish Hillary a Happy Birthday” ad captures this (see Figure 2). Not only is this appeal directed at supporters (what Republican will want to wish Hillary a happy birthday after all), it constructs a sense of what social identification with Hillary Clinton means: motherhood, family, warmth, care, and nurturing.

Figure 1: Hillary Clinton digital campaign advertisements

Figure 2: Hillary Clinton digital campaign advertisement

Source: Hillary for America Design 2016

Micro-targeting is also about the marking of difference. This is, perhaps, the most common trope in micro-targeted digital campaign ads. Campaigns look to not only establish the cultural meaning of their candidates and supporters, but also that of their opposition (Alexander, 2010). Donald Trump’s ads during the 2016 election reflected his rhetoric from the campaign trail in stressing themes of safety and security, in addition to the need to draw boundaries around civic incorporation (i.e., who should be allowed to be a citizen). For Hillary Clinton, micro-targeted ads were celebrations of diversity and multi-culturalism, especially the empowerment of women and racial and ethnic minorities. Political advertisements attempt to connect the candidates they promote with the demographic and social groups they seek to represent (in the United States this is at times drawn on racial and ethnic terms: whites for Republicans and a more diverse coalition for Democrats, see the discussion in Grossmann & Hopkins, 2016, 43-45).

In this, micro-targeting reflects and reinforces political agonism, the clash of competing social groups, interests, and values. Through micro-targeting, candidates stake out their claim to be on the civil side of the moral binary of the political sphere and strive to paint their opponents as anti-civil (Alexander, 2010). More colloquially, micro-targeted advertisements offer the beautiful affirmation of our values and the sharp critique of those of our opponents. Hillary Clinton’s campaign, for instance, clearly sought to portray Trump in terms of anti-civil racism, xenophobia, and sexism. And, the campaign used issues, such as abortion rights, and values, such as autonomy and choice, to build group identity and social solidarity around opposition to Trump: “Let’s stand together, join millions of women” (see Figure 3). This Facebook ad pits Clinton and her supporters against Trump and his supporters. Trump, in turn, combined nationalist and security appeals with an implicit construction of the American body politic in white identity terms (Figure 4). These ads capture the reality that political conflict is not only inevitable, but necessary: there are opposing views in politics on fundamental questions such as life, autonomy, and country. The audiences for these ads are not being presented with information to help them make up their own minds, they are being invited into a political struggle with clear opposing worldviews and moral values (see Figure 5). This is why mobilisation ads are directed towards identity-congruent audiences.

Figure 3: Hillary Clinton Facebook advertisement

Figure 4: Donald Trump digital advertisement

Source: Democracy in Action

Figure 5: Anti-Trump Hillary Clinton digital advertisements

Source: Hillary for America Design 2016

In these advertisements, it is also clear that micro-targeted ads present life as it should become, or as it must not become, linking the preferred candidate and political party with a civil vision of the future and the opposition with an anti-civil vision of the future, to use Alexander’s (2010) framework. As an example, for Ted Cruz (see Figure 6), the opposing side wants to infringe on the Bill of Rights, the fundamental liberty of Americans to defend their lives, liberties, families, and properties. Candidates run these issue ads to stake out their stance on the conflicting values, visions of the good life, plans for the future, and ends that are desirable in politics – whether it is embracing the freedom and security of gun rights for American Republicans or autonomy and choice in the context of reproductive rights for Democrats. These appeals are designed to mobilise the committed around the candidate’s vision of America’s past and future – they are designed for a world where we are sure of who we are and committed to our values and the ends we pursue.

Figure 6: Ted Cruz digital campaign advertisement

Source: Democracy in Action

Conclusion: democratic anxieties

I believe that there is such democratic anxiety about micro-targeting because citizens are supposed to be independent, autonomous, and rational. Micro-targeted advertising works to reinforce group identities and solidarity, mobilise partisans, and further the clash of political values. These things are all suspect from the perspective of the powerful and potent “folk theory” of democracy, as Achen and Bartels phrase it. As these realists argue, however, it’s far better to grapple with the reality of group-based democracy, with its attendant ingrained social allegiances and conflicts over values and power, rather than wishing for a transcendent and pure form of democracy without politics. These authors argue that we need to make peace with conflictual and competitive forms of group-based and pluralistic democracy premised on institutionally organised opposition. As Achen and Bartels (2015, 318) conclude:

Freedom is to faction what air is to fire, Madison said. But ordinary citizens often dislike the conflict and bickering that comes with freedom. They wish their elected officials would just do the people’s work without so much squabbling amongst themselves. They dislike the compromises that result when many different groups are free to propose alternative policies, leaving politicians to adjust their differences. Voters want “a real leader, not a politician,” by which they generally mean that their own ideas should be adopted and other people’s opinions disregarded, because views different from their own are obviously self-interested and erroneous. To the contrary, politicians with vision who are also skilled at creative compromise are the soul of successful democracy, and they exemplify real leadership.

My own view is that micro-targeting comes in the necessary service of this “conflict and bickering”. At its normative best, micro-targeting strengthens the hands of opposing factions, enabling them to identify and mobilise partisans to their cause, providing them with resources in terms of boots on the ground and money in the coffers. When opposing politicians and parties square off, they carry these resources into battle trying to advance their agendas or win concessions for their side. Compromise may be harder in a world of stronger factions, their hands steadied by the resources that micro-targeting can deliver, but that does not make compromise any less necessary or essential.

On the other hand, there are reasons for democratic concern about micro-targeting, but they look a bit different from narratives about public manipulation. Schudson (1986, 232) concludes that “advertising does not make people believe in capitalist institutions or even in consumer values, but so long as alternative articulations of values are relatively hard to locate in the culture, capitalist realist art will have some power.” I suspect that the same is true of political micro-targeting. The cultural power of political micro-targeting, but also political advertising more generally, lies in its creation of a set of ready-to-hand representations of democracy that citizens can express easily and fall back on. Taken to its extreme in a polarized political climate, micro-targeting can work to undermine the legitimacy of conflicts over opposing values and claims in democratic life. For example, in an undemocratic political culture micro-targeting can portray the other side as crooked and dangerous to the polity, political compromise as selling out, political expertise and representation as not to be trusted, and partisans’ own beliefs and identities as the only legitimate ones, not simply those among many in a pluralistic democracy. Micro-targeting also melds symbolic and social power in new ways, culturally legitimating and furthering the fortunes of autonomous and independent candidates, divorced from their parties and taking their appeals directly to voters (see Hersh, 2017).

This paper is part of 'A Manchurian candidate or just a dark horse? Towards the next generation of political micro-targeting research’, a Special issue of the Internet Policy Review.

Introduction

Political campaigns in the United States have employed digital technologies for more than a decade, developing increasingly sophisticated tools and techniques during each election cycle, as “computational politics” have become standard operating procedure (Tufecki, 2014; Kreiss, 2016). However, the most recent election marked a critical turning point, as candidates, political action committees, and other interest groups were able to take advantage of significant breakthroughs in data-driven marketing techniques, such as cross-device targeting, developed since the previous presidential election (“Bernie Sanders”, 2016; Edelman Digital, 2016). Electoral politics has now become fully integrated into a growing, global commercial digital media and marketing ecosystem that has already transformed how corporations market their products and influence consumers (Chahal, 2013; LiveRamp, 2015; Rubinstein, 2014; Schuster, 2015).The strategies, technologies, and tools of digital political marketing are more complex and far-reaching than anything we have seen before, with further innovations already underway (WARC, 2017). But because most commercial and political digital operations take place below the radar, they are not fully understood by the public. 1

In the following pages, we briefly describe the growth and maturity of digital marketing, highlighting its basic features, key players, and major practices. We then document how data-driven digital marketing has moved into the centre of American political operations, along with a growing infrastructure of specialised firms, services, technologies and software systems. We identify the prevailing digital strategies, tactics, and techniques of today’s political operations, explaining how they were employed during the most recent US election cycle. Finally, we explore the implications of their use for democratic discourse and governance, discussing several recent policy developments aimed at increasing transparency and accountability in digital politics.

Our research for this paper draws from our extensive experience tracking the growth of digital marketing over the past two decades in the United States and abroad, monitoring and analysing key technological developments, major trends, practices and players, and assessing the impact of these systems in areas such as health, financial services, retail, and youth (Chester, 2007; Montgomery, 2007; Montgomery, Chester, & Kopp, 2017). During the 2016 US presidential election, we monitored commercial digital advertising and data use by candidates, parties and special interest groups across the political spectrum. We collected examples of these ads, along with technical and market impact information from the developers of the applications. We also reviewed trade journals, research reports, and other industry documents, and attended conferences that were focused on digital technologies and politics. In the process, we identified all of the major providers of political digital data targeting applications (e.g., Google, Facebook, data clouds, ad agencies) and analysed all their key materials and case studies related to their 2016 operations. The source for much of this work was our ongoing gathering and analysis of cross-sectional commercial digital marketing practices worldwide.

Marriage of politics and commerce

Since the mid-20^th century, advertising has been an increasingly powerful and pervasive presence in US political campaigns, as a growing cadre of ad agencies, public relations firms, and consultants perfected the use of opinion polls, focus groups, and psychographics to reach and influence voters through radio, television, direct mail, and other media outlets (A. Jamieson, 2016; K. H. Jamieson, 1996; Sabato, 1981). With the rise of the internet, campaign operatives began to harness digital technologies and tools to mobilize voter turnout, engage young people, raise money, and support grassroots ground operations (Karpf, 2016; Kreiss, 2016; Tufecki, 2014). Both major political parties in the United States developed large, sophisticated data and digital operations (Kreiss, 2016).

Many of the digital strategies, tools, and techniques employed in the 2016 election were initially developed, deployed, tested, and refined by the commercial sector (Tufecki, 2014).Since its origins in the mid-1990s, digital marketing has operated with a core business model that relies on continuous data collection and monitoring of individual online behaviour patterns (Montgomery, 2011). This system emerged in the United States amid a political culture of minimal government interference, and within a prevailing laissez-faire ethos regarding the internet and new technologies (Barlow, 1996). In the earliest days of the “dot-com boom”, a strong political alliance was forged between the digital media companies and their partners in the advertising and media business, enabling the nascent industry to effectively ward off any attempts to restrain its business operations through privacy regulation or other public policies (Solon & Siddiqui, 2017). As a consequence, the advertising industry played a central role in shaping the operations of platforms and applications in the digital media ecosystem. Digital marketing is now well established and thriving, with expenditures reaching nearly $72.5bn in 2016 for the US alone, and worldwide spending predicted to reach more than $223bn this year (eMarketer, 2017; IAB, n.d.-d).

Ongoing innovations over the years have increased the capacity of data and digital marketing applications. Data collection, analysis, and targeting were further woven into the daily lives of consumers with the rise of social media platforms and mobile devices. Because of the unique role that they play in users’ lives, these platforms are able to sweep up enormous amounts of information, including not only what users post about themselves, but also what is collected from them throughout their daily activities (Smith, 2014). A growing arsenal of software and analytic tools has enhanced the ability of digital media companies and their advertisers to glean valuable insights from the oceans of data they generate (Smith, 2014). Predictive analytics introduced an expanded set of tools for scoring, rating, and categorising individuals, based on an increasingly granular set of behavioural, demographic, and psychographic data (“What is Predictive Intelligence”, 2017). US digital marketers have helped popularise and spur the successful adoption of digital advertising platforms and applications in nearly every geographical location with an internet connection or a link to a mobile device (IAB, n.d.-c). Google, Facebook, and other major players in the digital marketing industry have also developed a global research infrastructure to allow them, and especially their major advertising clients, to make continuous improvements in reaching and influencing the public, and to measure with increasing accuracy the success of their efforts (Facebook IQ, n.d.-a). These developments have created what some observers have called the “surveillance economy” (Singer, 2012).

The growth of data-driven political marketing

Though political campaigns have employed micro-targeting techniques—which use an array of personalised and other data sets and marketing applications to influence the actions of individuals—during the last several election cycles, recent technological innovations and industry advances have created a much more robust system than what was in place in 2012 (IAB, n.d.-b; Rubinstein, 2014). For years, political campaigns have been able to combine public voter files with commercial information from data brokers, to develop detailed and comprehensive dossiers on American voters (Rubinstein, 2014). With recent advances in the advertising technology and data industries, they can now take advantage of a growing infrastructure of specialty firms offering more extensive resources for data mining and targeting voters. Among the new entities are data marketing clouds. Developed by well-known companies such as Adobe, Oracle, Salesforce, Nielsen, and IBM, these clouds sell political data along with an exhaustive amount of detailed consumer information for each potential target, including, for example, credit card use, personal interests, consumption patterns, and TV viewing patterns (Salesforce DMP, 2017).

Some of these massive cloud services also operate what has become a new and essential component for contemporary digital targeting—the data management platform (DMP) (Chavez, 2017). DMPs provide marketers with “centralized control of all of their audience and campaign data” (BlueKai, 2011). They do this by collecting and analysing data about individuals from a wide variety of online and offline sources, including first-party data from a customer’s own record, such as the use of a supermarket loyalty card, or their activities captured on a website, mobile phone, or wearable device; second-party data, information collected about a person by another company, such as an online publisher, and sold to others; and third-party data drawn from thousands of sources, comprising demographic, financial, and other data-broker information, including race, ethnicity, and presence of children (O’Hara, 2016). All of this information can be matched to create highly granular “target audience segments” and to identify and “activate” individuals “across third party ad networks and exchanges”. DMPs are quickly becoming a critical tool for political campaigns (Bennett, 2016; Kaye, 2016, July; Regan, J., 2016).

Facebook and Google now play a central role in political operations, offering a full spectrum of commercial digital marketing tools and techniques, along with specialied ad “products” designed for political use (Bond, 2017). Not surprisingly, these companies have also made generating revenues from political campaigns an important “vertical” category within their ad business (Facebook, n.d.-d; Facebook IQ, n.d.-b; Stanford, 2016). Facebook’s role in the 2016 election was particularly important. With users required to give their real names when they sign up as members, Facebook has created a powerful “identity-based” targeting paradigm, enabling political campaigns to access its more than 162 million US users and to target them individually by age, gender, congressional district, and interests (Facebook, n.d.-b).Its online guide for political campaign marketing urges political campaigns to use all the social media platform tools it makes available to advertisers—including through Instagram and other properties—in order to track individuals, capture their data through various “lead-generation” tactics, and target them by uploading voter files and other data (Facebook, n.d.-a-c-f). The company also employs teams of internal staff aligned with each of the major political parties to provide technical assistance and other services to candidates and their campaigns (Chester, 2017; Kreiss & Mcgregor, 2017). Google heavily promoted the use of YouTube, as well as its other digital marketing assets, during the 2016 US election, reaching out to both major political parties (YouTube, 2017).

The growth and increasing sophistication of the digital marketplace has enhanced the capacities of political campaigns to identify, reach, and interact with individual voters. Below we identify seven key techniques that are emblematic of this new digital political marketing system, providing brief illustrations of how they were employed during the 2016 election.

Cross-device targeting

Getting a complete picture of a person’s persistent “identity” through an “identity-graph” has become a key strategy for successfully reaching consumers across their “omnichannel” experience (use of mobile, TV, streaming devices, etc.) (Winterberry Group, 2016). “Cross-device recognition” allows marketers to determine if the same person who is on a social network is also using a personal computer and later watching video on a mobile phone. Through data “onboarding,” a customer record that may contain a physical and email address is linked through various matching processes, associating it with what is believed to be that individual’s online identification—cookies, IP addresses, and other persistent identifiers (Levine, 2016). Cross-device targeting is now a standard procedure for political initiatives and other campaigns. Voter files are uploaded into the onboarding process, enabling the campaigns to find their targets on mobile devices and at specific times when they may be more receptive to a message (Kaye, 2016, April; L2, n.d.-b). Such granularity of information also enables a more tailored advertisement—so-called “dynamic creative”—which can be changed over time to “deliver very specific messaging” to individuals (Schuster, 2015). Leading cross-device marketing company Drawbridge offered a suite of election services in 2016 that provided campaigns a number of ways to impact voters, including through “Voter-Centric Cross Device Storytelling”, “Political Influencer Identification”, and via “Real-Time Voter Attribution Measurement” (Drawbridge, n.d.).

Programmatic advertising

Programmatic advertising refers to new automated forms of ad buying and placement on digital media using computer programmes and algorithmic processes to find and target a customer wherever she goes. The process can also involve real-time “auctions” that occur in milliseconds in order to “show an ad to a specific customer, in a specific context” (Allen, 2016). The use of programmatic advertising was one of the major changes in political campaign digital operations between 2012 and 2016—“the first time in American History”, according to one ad company, “that such precise targeting has ever been made available at such great scale” (Briscoe, 2017; Kaye, 2015). Programmatic advertising has itself grown in its capabilities to reach individuals, taking advantage of new sources of data to reach them on all of their devices (Regan, T., 2016). In 2016, for example, global ad giant WPP’s Xaxis system—“the world’s largest programmatic and technology platform”—launched “Xaxis Politics”. Capable of “reaching US voters across all digital channels”. the system is said to “segment audiences by hundreds of hot button issues as well as by party affiliation”, including via “real-time campaigns tied to specific real-world events” (Xaxis, 2015). Candidates were able to use the services of a growing list of companies, including Google, Rubicon, AOL, PubMatic, Appnexus and Criteo, that offered programmatic advertising platforms (“Political Campaigns”, 2016; Yatrakis, 2016).

Lookalike modelling

The use of big data analytics enables marketers to acquire information about an individual without directly observing behaviour or obtaining consent. They do this by “cloning” their “most valuable customers”in order to identify and target other prospective individuals for marketing purposes (LiveRamp, 2015). For example, Stirista (n.d.), a digital marketing firm that also serves the political world, offers lookalike modelling to identify people who are potential supporters and voters. The company claims it has matched 155 million voters to their “email addresses, online cookies, and social handles”, as well as “culture, religion, interests, political positions and hundreds of other data points to create rich, detailed voter profiles”. Facebook offers a range of lookalike modelling tools through its “Lookalike Audiences” ad platform. For example, Brad Parscale, the Trump campaign’s digital director, used the Lookalike Audiences ad tool to “expand” the number of people the campaign could target (Green & Issenberg, 2016). Facebook’s “Custom Audiences” product, similarly, enables marketers to upload their own data files so they can be matched and then targeted to Facebook users (Facebook, n.d.-e).

Geolocation targeting

Mobile devices continually send signals that enable advertisers (and others) to take advantage of an individual’s location—through the phone’s GPS (global positioning system), Wi-Fi, and Bluetooth communications. All of this can be done with increasing speed and efficiency. Through a host of new location-targeting technologies, consumers can now be identified and targeted wherever they go, while driving a car, pulling into a mall, or shopping in a store (Son, Kim, & Shmatikov, 2016). A complex and growing infrastructure of geolocation-based data-marketing services has emerged, with specialised mobile data firms, machine-learning technologies, measurement companies, and new technical standards to facilitate on-the-go targeting (Warrington, 2015). The use of mobile geo-targeting techniques played a central role in the 2016 election cycle, with a growing number of specialists offering their services to campaign operatives. For example, L2 (n.d.-a) made its voter file, along with HaystaqDNA modelling data, available for mobile device targeting, offering granular profile data on voters based on their interest in such contested topics as gun laws, gay marriage, voter fraud, and school choice, among others. Conde Nast’s Advance Publications’ election advertising worked with campaigns to append geo-location, profile data, and buying behaviour “to sculpt a very specific voter profile and target down to few hundred readers in a given geo location” (Ellwanger, 2016).

Online video advertising

Digital video, via mobile and other devices, is perceived as a highly effective way of delivering emotional content on behalf of brands and marketing campaigns (IAB, n.d.-a). There are a variety of online video ad formats that provide both short- and long-form content, and that work well for political and other marketing efforts. Progressive political campaign company Revolution Messaging, which worked for the Sanders campaign, developed what it calls “smart cookies” that it says take video and other ad placement “to the next level, delivering precision and accuracy” (Revolution Messaging, n.d.). Google’s YouTube has become a key platform for political ads, with the company claiming that today, voters make their political decisions not in “living rooms” in front of a television but in what it calls “micromoments” as people watch mobile video (DoubleClick, n.d.). According to the company’s political ad services research, mobile devices were used in nearly 60 percent of election-related searches during 2016. Content producers (which it calls “Creators”) on YouTube were able to seize on these election micro-moments to influence the political opinions of potential voters 18-49 (“Letter from the Guest Editors,” 2016).

Targeted TV advertising

Television advertising, which remains a linchpin of political campaign strategy, is undergoing a major transformation, as digital technologies and “addressable” set-top boxes have changed cable and broadcast TV into powerful micro-targeting machines capable of delivering the same kinds of granular, personalised advertising messages to individual voters that have become the hallmark of online marketing. Political campaigns are in the forefront of using set-top box “second-to-second viewing data”, amplified with other data sources, such as “demographic and cross-platform data from a multitude of sources” via information brokers, to deliver more precise ads (Fourthwall Media, n.d.; Leahey, 2016; NCC Media, n.d.). NCC Media, the US cable TV ad platform owned by Comcast, Cox, and Spectrum, provided campaigns the ability to target potential voters via the integration of its set-top box viewing information with voter and other data from Experian and others (Miller, 2017). Deals between TV data viewing companies and organisations representing both Republican- and Democratic-leaning groups brought the “targeting capabilities of online advertising to TV ad buys…bringing what was once accessible only to large state-wide or national campaigns to smaller, down-ballot candidates”, explained Advertising Age (Delgado, 2016).

Psychographic, neuromarketing, and emotion-based targeting

Psychographics, mood measurement, and emotional testing have been used by advertisers for many decades, and have also been a core strategy in political campaign advertising (Key, 1974; Packard, 2007; Schiller, 1975). The digital advertising industry has developed these tools even further, taking advantage of advances in neuroscience, cognitive computing, data analytics, behavioural tracking, and other recent developments (Crupi, 2015). Granular-based messages that trigger a range of emotional and subconscious responses, to better “engage” with individuals and deepen relationships with commercial brands, have become part of the DNA of digital advertising (McEleny, 2016). Facebook (2015), Nielsen, and most leading brands use “neuromarketing” services worldwide, which utilise neuroscience tools to determine the emotional impact of advertising messages. There is a growing field, recently promoted by Google, of “Emotion Analytics” that takes advantage of “new types of data and new tracking methods” to help advertisers “understand the impact of campaigns—and their individual assets—on an emotional level…” (Kelshaw, 2017). Scholars have identified that the use of “psychological targeting” in advertising enables the influencing of large groups of people by “tailoring persuasive appeals to the psychological needs” of the specific audiences (Matz, et al, 2017). Data company Experian Marketing Services for political campaigns offered data that weaved together “demographic, psychographic and attitudinal attributes” to target voters digitally. Experian claims its data enables campaigns to examine a target’s “heart and mind” via attributes related to their “political persona” as well as “attitudes, expectations, behaviours, lifestyles, purchase habits and media preferences (Experian, 2011, 2015). One of the most well publicised and controversial players in the 2016 election was Cambridge Analytica (CA), a prominent data analytics and behavioural communications firm that claimed to be a key component in Donald Trump’s victorious campaign. The company used a “five-factor personality model” aimed at determining “the personality of every single adult in the United States of America” (Albright, 2016; Kranish, 2016).Known as OCEAN, the model rated individuals based on five key traits: openness, conscientiousness, extroversion, agreeableness, and neuroticism. Drawing from digital data, voter history, and marketing resources supplied by leading companies, including Acxiom, Experian, Nielsen, GOP firm Data Trust, Aristotle, L2, Infogroup, and Facebook, CA was able to develop an “internal database with thousands of data points per person”. Its research also identified key segments that were considered “persuadable”, and shaped the advertising content placed “across multiple digital channels (with the most effective ads also appearing on television) (Advertising Research Foundation, 2017; Nix, 2016). The strategy was based on developing messages that were tailored to the vulnerabilities of individual voters (Nix, 2016; Schwartz, 2017). CA has become the subject of much scrutiny and debate, and itself has made conflicting claims, with critics raising concerns over its techniques and expressing scepticism about the extent of its impact (Confessore & Hakim, 2017; Karpf, 2017). However, the company’s work was sufficiently convincing to the leading advertising industry research organisation, the Advertising Research Foundation (2017, March), that it honoured the firm with a “Gold” award in 2017 under its “Big Data” category.

Discussion

The above description provides only a brief overview of the data-driven marketing system that is already widely in use by candidate and issue campaigns in the United States. The increasingly central role of commercial digital marketing in contemporary political campaigns is reshaping modern-day politics in fundamental ways, altering relationships among candidates, parties, voters, and the media. We acknowledge that digital technologies have made important positive contributions to the vibrancy of the political sphere, including greatly expanding sources of news and information, significantly increasing opportunities for citizen participation, and empowering people from diverse backgrounds to form coalitions and influence policy. The same tools developed for digital marketing have also helped political campaigns substantially improve voter engagement, enhance their capacities for “small-donor” fundraising, and more efficiently generate turnout (Moonshadow Mobile, n.d.; Owen, 2017). However, many of the techniques we address in this paper raise serious concerns—over privacy, discrimination, manipulation, and lack of transparency.

Several recent controversies over the 2016 election have triggered greater public scrutiny over some of the practices that have become standard operating procedure in the digital media and marketing ecosystem. For example, “fake news” has a direct relationship to programmatic advertising, the automated system of “intelligent” buying and selling of individuals and groups (Weissbrot, 2016). These impersonal algorithmic machines are focused primarily on finding and targeting individual consumers wherever they are, often with little regard for the content where the ads may appear (Maheshwari & Isaac, 2016). As a consequence, in the middle of the 2016 election, many companies found themselves with ads placed on “sites featuring pornography, pirated content, fake news, videos supporting terrorists, or outlets whose traffic is artificially generated by computer programs”, noted the Wall Street Journal (Nicas, 2016; Vranica, 2017). As a major US publisher explained in the trade publication Advertising Age,

Programmatic’s golden promise was allowing advertisers to efficiently buy targeted, quality, ad placements at the best price, and publishers to sell available space to the highest bidders…. What was supposed to be a tech-driven quality guarantee became, in some instances, a “race to the bottom” to make as much money as possible across a complex daisy chain of partners. With billions of impressions bought and sold every month, it is impossible to keep track of where ads appear, so “fake news” sites proliferated. Shady publishers can put up new sites every day, so even if an exchange or bidding platform identifies one site as suspect, another can spring up (Clark, 2017).

Criticism from news organisations and civil society groups, along with a major backlash by leading global advertisers, led to several initiatives to place safeguards on these practices (McDermott, 2017; Minsker, 2017). For example, in an effort to ensure “brand safety”, leading global advertisers and trade associations demanded changes in how Google, Facebook and others conduct their data and advertising technology operations. As a consequence, new measures have been introduced to enable companies to more closely monitor and control where their ads are placed (Association of National Advertisers, 2017; Benes, 2017; IPA, 2017; Johnson, 2017; Liyakasa, 2017; Marshall, 2017; Timmers, 2015).

The Trump campaign relied heavily on Facebook’s digital marketing system to identify specific voters who were not supporters of Trump in the first place, and to target them with psychographic messaging designed to discourage them from voting (Green & Issenberg, 2016). Campaign operatives openly referred to such efforts as “voter suppression” aimed at three targeted groups: “idealistic white liberals, young women and African Americans”. The operations used standard Facebook advertising tools, including “custom audiences” and so-called “dark posts”—“nonpublic paid posts shown only to the Facebook users that Trump chose” with personalised negative messages (Green & Issenberg, 2016). Such tactics also took advantage of commonplace digital practices that target individual consumers based on factors such as race, ethnicity, and socio-economic status (Google, 2017; Martinez, 2016; Nielsen, 2016). Civil rights groups have had some success in getting companies to change their practices. However, for the most part, the digital marketing industry has not been held sufficiently accountable for its use of race and ethnicity in data-marketing products, and there is a need for much broader, industry-wide policies.

Conclusion

Contemporary digital marketing practices have raised serious issues about consumer privacy over the years (Schwartz & Solove, 2011; Solove & Hartzog, 2014). When applied to the political arena, where political information about individuals is only one of thousands of highly sensitive data points collected and analysed by the modern machinery of data analytics and targeting, the risks are even greater. Yet, in the United States, very little has been done in terms of public policy to provide any significant protections. In contrast to the European Union, where privacy is encoded in law as a fundamental right, privacy regulation in the US is much weaker (Bennett, 1997; Solove & Hartzog, 2014; U.S. Senate Committee on Commerce, Science, and Transportation, 2013). The US is one of the only developed countries without a general privacy law. As a consequence, except in specific areas, such as children’s privacy, consumers in the US enjoy no significant data protection in the commercial marketplace. In the political arena, there is even less protection for US citizens. As legal scholar Ira S. Rubinstein (2014) explains, “the collection, use and transfer of voter data face almost no regulation”. The First Amendment plays a crucial role in this regard, allowing the use of political data as a protected form of speech (Persily, 2016).

The political fallout over the how Russian operatives used Facebook, Twitter, and other sites in the 2016 presidential campaign has triggered unprecedented focus on the data and marketing operations of these and other powerful digital media companies. Lawmakers, civil society, and many in the press are calling for new laws and regulations to ensure transparency and accountability for online political ads (“McCain, Klobuchar & Warner Introduce Legislation”, 2017). The U.S. Federal Election Commission, which regulates political advertising, has asked for public comments on whether it should develop new disclosure rules for online ads (Glaser, 2017). In an effort to head-off regulation, both Facebook and Twitter have announced their own internal policy initiatives designed to provide the public with more information, including what organisations or individuals paid for political ads and who the intended targets were. These companies have also promised to establish archives for political advertising, which would be accessible to the public (Falck, 2017; Goldman, 2017; Koltun, 2017). The US online advertising industry trade association is urging Congress not to legislate in this area, but to allow the industry to develop new self-regulatory regimes in order to police itself (IAB, 2017). However, relying on self-regulation is not likely to address the problems raised by these practices and may, in fact, compound them. Industry self-regulatory guidelines are typically written in ways that do not challenge many of the prevailing (and problematic) business practices employed by their own members. Nor do they provide meaningful or effective accountability mechanisms (Center for Digital Democracy, 2013; Gellman & Dixon, 2011; Hoofnagle, 2005). It remains to be seen what the outcome of the current policy debate over digital politics will be, and whether any meaningful safeguards emerge from it.

While any regulation of political speech must meet the legal challenges posed by the First Amendment, limiting how the mining of commercial data can be used in the first place can serve as a critically important new electoral safeguard. Advocacy groups should call for consumer privacy legislation in the US that would place limits on what data can be gathered by the commercial online advertising industry, and how that information can be used. Americans currently have no way to decide for themselves (such as via an opt-in) whether data collected on their finances, health, geo-location, as well as race or ethnicity can be used for digital ad profiling. Certain online advertising practices, such as the use of psychographics and lookalike modelling, also call for rules to ensure they are used fairly.

Without effective interventions, the campaign strategies and practices we have documented in this paper will become increasingly sophisticated in coming elections, most likely with little oversight, transparency, or public accountability. The digital media and marketing industry will continue its research and development efforts, with an intense focus on harnessing the capabilities of new technologies, such as artificial intelligence, virtual reality, and cognitive computing, for advertising purposes. Advertising agencies are already applying some of these advances to the political field (Facebook, 2016; Google, n.d.-a; Havas Cognitive, n.d.). Academic scholars and civil society organisations will need to keep a close watch on all these developments, in order to understand fully how these digital practices operate as a system, and how they are influencing the political process. Only through effective public policies and enforceable best practices can we ensure that digital technology enhances democratic institutions, without undermining their fundamental goals.

On the International Data Protection Day, 28 January 2019, Panoptykon Foundation filed complaints against Google and Interactive Advertising Bureau (IAB) Europe under the General Data Protection Regulation (GDPR) to the Polish Data Protection Authority (DPA), joining the ad auction complaints already being examined in the UK and in Ireland. The complaints are related to the functioning of online behavioural advertising (OBA) ecosystem, in particular the fact that massive amounts of personal data (including sensitive information) are leaked in the real time bidding process.

Context

Every time a user visits a website that is connected with the ad auction ecosystem, his personal data is broadcast in a bid request to tens or hundreds of companies. Bid requests contain categories that reflect what content this particular user was reading or watching, as well as his recent search query. These categories can be benign, such as “bowling” or “gadgets”, but also extraordinarily sensitive.

For example, one category is “IAB7-28 Incest/Abuse Support”. This could enable ad auction companies to target and profile a person as an incest or abuse victim. Other categories relate to sensitive and embarrassing health conditions, religious denomination, sexual orientation, etc. Google runs its own category list, which includes equally sensitive insights such as “eating disorders”, “left-wing politics”, or “scientology”.

Arguments

Complaints focus on the role of Google and the Interactive Advertising Bureau (IAB) as organisations that set standards for other actors involved in the OBA market. They should therefore be treated as data controllers responsible for GDPR infringements. Arguments used by Panoptykon are based on complaints filed in the UK and Ireland in September 2018 by Jim Killock of the Open Rights Group, Michael Veale of University College London, and Dr Johnny Ryan of Brave

Key facts and observations of the complaints:

• data shared by companies within the OBA ecosystem are not necessary for the purposes of serving targeting ads; in particular there are no legal grounds to share sensitive data of internet users, which – in accordance with taxonomies defined by IAB and Google – may be included in bid request;

• companies sharing data have no control over its further use by a potentially unlimited number of other actors that have access to real-time bidding software;

• users have no access to their data and no tools of controlling its further use by a (potentially unlimited) number of actors;

• those failures are not incidental because they result from the very design of the OBA ecosystem - lack of transparency and the concept of bid request, which, by definition, leads to data "broadcasting".

Responses

In response to the complaint, IAB made a point that it is not in control of personal data that flow through the ad auction ecosystem. Purportedly their role is limited to defining technical protocol, which other companies choose to implement, at their discretion. According to IAB it is the responsibility of these companies to ensure that whatever personal data they pass or receive via the bidstream complies with the particular restrictions of their jurisdiction.

Brave, ORG and Panoptykon are not convinced with this argument. It is true that the technical protocol defined by IAB does not force companies to share personal data of their users but the very task of this protocol is to enable and facilitate the flow of personal data. In fact, the only way for companies to benefit (economically) from joining the ad auction is to provide personal data, as defined by the protocol.

The IAB “AdCOM” specification suggests that personal identifiers “about the human user of the device” are “strongly recommended” to be involved in a bid request. Whilst not technically “required” as defined within the specification, the inclusion of data is said to be “recommended” in bid requests “when their omission would not break the object, but would dramatically diminish its value.”

IAB also purports that its taxonomy has nothing to do, as such, with processing sensitive data of individuals because its only task is to mark up content on the website. Claimants acknowledge that, indeed, IAB and Google taxonomy is used to categorise content. But – they argue in response – when labelled content is used by a person, certain labels stick to this particular user and become his personal data.

There is nothing wrong with categorising content. But while it is acceptable for a library to mark an area with the words "substance abuse", it would not be acceptable for a library to mark a person who enters that section with those words too. Online, these labels about what users read, watch, and listen to can stick to them for a long time.

Prior investigation and report

Prior to making complaints, Panoptykon carried its own investigation of the OBA ecosystem in Poland, which confirmed allegations made by Brave and ORG in their complaints, as well as Johnny Ryan's testimony. Between May and December 2018, Panoptykon sent a number of data access requests to various actors involved in the OBA ecosystem (including Google and leading data brokers) in order to check whether users are able to verify and correct their marketing profiles.

In most cases companies refused to provide personal data to users based on alleged difficulty with their identification.This argument - made by key players in the OBA ecosystem - confirms that it has been designed to be obscure. Key identifiers used by data brokers to single out users and target ads are not revealed to data subjects that are concerned. It is a "catch 22" situation that cannot be reconciled with GDPR requirements (in particular the principle of transparency).

Along with its complaints, Panoptykon published a report in Polish summarising its investigation of the OBA ecosystem, which included interviews with key actors operating on the Polish market, and evidence collected by sending data access requests.