The rise of populist radical right parties (PRRPs) is largely seen to have been triggered by a dealignment between voters and political elites and to have triggered an increasing supply of misinformation. Consequently, the German populist radical right party Alternative für Deutschland (AfD) is strongly associated with the notion that ‘fake news’ endanger democracy (Reuter et al., 2019). Whereas most studies focus on ‘fake news’ dynamics in social media, this contribution analyses one-to-one communication between politicians and constituents in an experimental setting. For this purpose, all 2,503 German parliamentarians of the federal political system receive an artificial inquiry including a piece of misinformation that circulates among the different electorates. The experiment reveals a division in political communication between established parties and the AfD. Politicians of established parties seem to be intrinsically driven by the ideal of a good democratic representative. They significantly abstain from tolerating misinformation, are more responsive and their overall communication behaviour is much less strategic. What is more, AfD parliamentarians are seven times as eager to tolerate ‘fake news’ – a number that becomes even higher when facing an immigration related piece of misinformation circulating within the populist radical right’s electorate. Finally, the AfD’s communication behaviour reflects their intention to sow distrust in the German society, economy, and administration.

1. Introduction

It is the nature of political competition that different parties stress different aspects of the same political issue. The recent debate on ‘fake news’ and ‘factfulness’, however, has raised awareness for a structural manipulation of potential voters’ opinions (Pennycook et al., 2018). Hence, when analysing US politics “do facts matter?” is the fundamental question asked by Hochschild and Einstein (2015). Anti-establishment populist radical right parties (PRRPs) – who claim to be particularly responsive to ‘the people’ (Mudde, 2007) – have been accused of strategically disregarding facts, spreading misinformation, and triggering a polarization of the entire political system (Bergmann, 2018; Castanho Silva, 2018; Runciman, 2018). These pieces of misinformation can contain complex conspiracy myths. However, ‘fake news’ also include rather simple falsifiable statements such as White House press secretary Sean Spicer’s pretension that Donald Trump was received by the largest inauguration crowd in the history of the United States (The Guardian, 2017), Boris Johnson’s famous Brexit campaign claim the UK would send 350 million Pound to the EU every week (Full Fact, 2017) or the German Alternative für Deutschland’s (AfD) exaggeration that the foreigners’ homicide rate in Germany had increased by 685 percent in 2018 (Tagesschau, 2019). Successful political campaigns manage to circumvent traditional gatekeepers by exploiting digital filter bubbles: Particularly, the radical right has become a super-spreader of misinformation and tolerates ‘fake news’ within their communities as it serves their political agenda (Allcott and Gentzkow, 2017; Cantarella et al., 2019; Guess et al., 2020). Additionally, for Germany Knuth and Mayr conclude: “Like Trump, the AfD lives on the decay of certainty” (own translation, 2020).

Although the ‘fake news’ literature is limited to the semi-public digital world mostly focusing on social media, unfiltered political communication also involves direct one-to-one voter-politician communication. Recently, these discourses have been analysed in innovative responsiveness experiments based on constituent inquiries: In the US, senators were found to strategically tailor their answers on a controversial issue congruent with the stated constituents’ preferences – although without explicitly spreading ‘fake news’ (Grose et al., 2015). Until today, this strand of literature leaves aside the special role, direct voter-politician communication plays for the radical right. Whereas PRRPs seem to be the driving force behind misinformation circulating in social media, it remains an open question whether the same is true for one-to-one communication between voters and politicians. The following analysis bridges this gap by carrying out a citizens’ inquiry experiment addressing the following research questions:

Does PRRP politicians’ responsiveness stand out in one-to-one communication with constituents? Or is political communication, spreading and tolerating misinformation more generally related to party specific issue salience and positioning?

The German political system represents a fruitful example for such a responsiveness misinformation communication experiment focussing on PRRP parliamentarians. On the one hand, in times of COVID-19 several German parliamentarians have stressed the importance of answering citizens’ inquiries remotely (Tagesspiegel, 2020). On the other hand, the far right AfD regularly claims to be particularly responsive and to represent ‘the true people’ (“wir sind das Volk”) (Geiges, 2018; Siri and Lewandowsky, 2019) while being criticized of spreading misinformation by its competitors and the media (br, 2020; mdr, 2019; ZeitOnline, 2019).

What is more, 89 AfD parliamentarians on the national and 236 AfD parliamentarians on the state level represent a significant number of potential respondents. At the same time, the six large factions present in the German parliament allow a gradual responsiveness analysis depending on the respective party’s positioning and issue salience. Particularly, the AfD’s cultural antagonist, the Green Party enables a viable comparison of political communication in constituent inquiries – with 67 parliamentarians on the national and 266 parliamentarians on the state level.

The present analysis builds a bridge between the PRRP misinformation literature that solely focusses on social media and the experimental approaches that miss to zoom in on the populist radical right behaviour in answering citizens inquiries: A total population experiment of one-to-one citizens-parliamentarians communication of all 2.580 political representatives of a federal political system is carried out. The concept of ‘fake news’ is operationalized for the AfD’s and the Green Parties’ core issues – immigration and climate change – and a control topic based on misinformation that actually circulate within the electorate. Finally, the experiment fills a pressing gap in the literature by analysing the response behaviour in direct citizens-parliamentarians communication when being confronted with misinformation.

The remainder of this paper is structured as follows. The following section gives an overview over the ‘fake news’ and responsiveness debate in the context of the right-wing populist rise. A special focus is laid on experimental approaches to explore politicians’ communication behaviour and the mechanisms why misinformation would be spread or tolerated when being directly addressed by constituents. The third section explains the empirical strategy and the experimental design that is chosen to pursue the research question. The fourth section explains the results and carves out whether the AfD plays a special role in spreading misinformation and in the voter-politician responsiveness. A final section concludes with the most important findings. Important ethical considerations are summarised in Appendix 1.

2. Literature review

2.1 Tailoring the truth on the populist right: The German case

It is a well-established finding that political competition contains more than simple differences in parties’ issue positioning. When political actors jockey for power, language plays a fundamental role (Korte, 2002). Which political voices are heard, which topics can be introduced to the political arena (agenda setting) and how issues are framed is crucial for political power structures and finally for election outcomes (Klüser and Radojevic, 2016). Scholars have recently focused on the art of narrative storytelling in political competition: Namely, the importance of weaving a political or economic agenda into powerful emotionally anchored narrations (Koschorke, 2012; Shiller, 2019). The populist radical right employs such communication strategies to break through the “conspiracy of silence” (Arzheimer, 2009) that was established around them. By introducing extreme language and new anti-immigration narratives, PRRPs manage to gain centre stage in the political discourse, finally triggering a strongly polarized political system (Castanho Silva, 2018).

The understanding of political competition as a struggle over genuinely normative narratives puts the lively debate on the rise of the radical right into perspective. Indeed, electoral success by PRRPs has been answered by the established political powers ringfencing the newcomers’ extreme positioning and language in a “cordon sanitaire” (Downs, 2002; Heinze, 2018; Mudde, 2007). By making sure PRRPs are considered disreputable, established players intent to prevent that gatekeepers such as media representatives allow the right-wingers from feeding their messages into the established public space.² Such a strategy is fundamentally put into question by the mass use of social media, opening a new channel to directly address the electorate that is both cheaper and more lucrative than large scale campaigns:³ By blocking radical ‘insurgent’ politicians from traditional media, the radicals are virtually forced into social media campaigns. Successful social media strategies are built on regularly causing sensations by extreme exaggerations or deliberate provocations that traditional media inevitably must pick up (Manow, 2020). In this context it is argued that polarization itself has become a political business model (Taibbi, 2019): By stressing negative (fake) stories on immigration, agitating radical right candidates have taken advantage of vicious circles of polarization and (social) media attention (Berning et al., 2019; Gerstlé and Nai, 2019; Mounk, 2018; Reuning and Dietrich, 2018; Schmidt, 2020). Interestingly, the same logic – however, based on different issues – has been detected in campaigns of radical left-wing candidates such as of Jeremy Corbin (Prince, 2016) or Bernie Sanders (Penney, 2017).

The rise of such radical politicians and their (social) media coverage is often associated with the rise of misinformation – manipulative “by design” (Gelfert, 2018) – up to the proclamation of an age of “post-truthism” (The Economist, 2016). ‘Fake news’, the post-truth currency, are defined as “false stories that appear to be news, spread on the internet or using other media, usually created to influence political views or as a joke” (Cambridge Dictionary, 2020). Such a definition goes beyond ‘normal’ tailoring of the truth in narrative political storytelling (Gelfert, 2018) and resides somewhere in the “grey zone between advertising and propaganda” (Tandoc et al., 2018, p. 10).⁴

In Germany ‘fake news’ are largely attributed to the populist radical right-wing party AfD. The German PRRP employs an anti-elite rhetoric, promotes a traditional way of life, stages itself as the defender of a homogeneous, nativist nation state, opposes immigration and denies that climate change resulted from human activities (Arzheimer and Berning, 2019; Jacob et al., 2020; Schmitt-Beck, 2017) (see Figure 1 for an overview over positioning and salience). Mirroring its character as an anti-immigration party, the party’s social media activity shows the typical issue focus of migration, Europe, and Islam (Arzheimer and Berning, 2019). Moreover, typically for a PRRP, during the 2017 election campaign the party threatened to sue public television broadcasters for being underrepresented in major political talk shows (Focus, 2017). Furthermore, the AfD dominates online communication among German parties and manages to circumvent potentially dividing topics by aggressively focusing on anti-immigration positions (Serrano et al., 2019). Unsurprisingly, ‘fake news’ trackers reveal that seven of the ten most spread cases in the German federal election campaign 2017 have been shared by the German PRRP (Sängerlaub et al., 2018), nourishing anti-immigration sentiments by including stories of immigrants as rapists and looters (Siri and Lewandowsky, 2019). Also, more subtle misinformation such as the underestimation of refugees’ education levels have been discussed controversially during the federal election campaign 2017 as several AfD representatives circulated an article claiming only one third of immigrants graduated from high school, whereas the true share lies much higher (ZeitOnline, 2017).⁵ In fact, the AfD is strongly attributed to the rise of ‘fake news’ in Germany and the increasing perception that the spread of misinformation represents a threat for democracy in Germany (Reuter et al., 2019). In this context it is important to note that the nature of social media enables AfD politicians to spread anti-immigration content from dubious sources by liking and sharing without necessarily making things up on their own.

On the other cultural end of the party system, the Green Party embraces a multi-cultural society, favours post-materialist values, advocates Germany as a country of immigration and stresses the German responsibility in the fight against climate change as their main topic (Bündnis 90/Die Grünen, 2020). Correspondingly, the Greens are positioned at the other extreme than the AfD in the policy dimensions of immigration and the environment (see Figure 1). In contrast to most European countries, the German Greens manage to politically cash in on their climate change issue-ownership – especially in the European election of 2019 (Probst, 2020). The party’s recent success is closely associated with rising issue salience of climate change in the context of the worldwide Fridays for Future grassroot demonstrations (Weyers et al., 2020). In the German climate change debate the Greens were accused by the AfD of “hysterical climate-crisis screeching” (own translation, Frindte and Frindte, 2020) as the right-wingers demanded “fact-based climate and energy politics” (own, translation, Deutscher Bundestag, 2018).

Figure 1: Issue positioning in the German party system; source: Chapel Hill Expert Survey (2020)

Hence, the AfD and the Greens promote opposing policy platforms in various topics – although the AfD’s environmental positioning is rather low in salience (see Figure 1). Their electoral success can be clearly attributed to a single-issue positioning: anti-immigration for the AfD; the fight against climate change for the Greens. Furthermore, AfD politicians regularly assert a particular responsiveness as only they would be the true spokesperson of ‘the people’ while being criticized of spreading misinformation. The Green Party successfully jumped on the popularity bandwagon of the anti-climate-change movement by demanding political responsiveness to the influential grassroot movement Fridays for Futures that had to be “taken serious” (own translation, Bundestagsfraktion Bündnis 90 / Die Grünen, 2019). Their ‘factfulness’ in the climate change debate has been questioned by the AfD.

2.2 New experimental evidence: responsiveness tests with parliamentarians

The discussed literature largely links the spread and tolerance of misinformation (by the radical right) to semi-public social media communication. However, unfiltered direct communication between politicians and voters could be equally exploited for such purpose – while flying under the radar of the public due to its lack of traceability. An important example of citizens getting in touch with their representatives at low-cost are constituency inquiries (“contacting”) (Teorell et al., 2006). For parliamentarians, inquiries from their constituency represent a comparably cheap and well-targeted chance to convince potential voters (Cain et al., 1984). More recently, digital one-to-one contacting of parliamentarians by voters has been analysed, as the mass use of internet communication lowered the threshold to get in touch with authorities even further. Already in 2004, members of the US Congress were bombarded with 200 million e-mails and letters (Fitch and Goldschmidt, 2005).

For reasons of data privacy protection, no such statistics are published or available upon request for Members of the German Bundestag.⁶ As in the US, the varieties of political communication and particularly the communication with their constituencies represents one of the core responsibilities of German MP offices (Bröchler and Elbers, 2001). Bundestag parliamentarians and their staff are understood to function as “political high-performance teams” (own translation, Lattrari, 2020) that efficiently deal with a great variety of different tasks. However, there is a large diversity concerning the level of specialization and distribution of tasks between MPs and their employees. For the political communication this means that although oftentimes citizen requests might be answered not by MPs themselves, citizens do not necessarily realize their inquiry has not been answered by staff members.⁷ It is important to note that parliamentarians in Germany stand on a level playing field in terms of funding, they are free to pick their own staff and design their own communication strategy. The Bundestag offers a monthly budget of 19.913 Euros for each parliamentarian to be spent on their employees financing the salaries of a total staff of 4.500 people (Deutscher Bundestag, 2015). It remains in the control of individual parliamentarians to devote resources to answer citizen inquiries.⁸

Somewhat surprisingly given the vast (digital) engagement between representatives and their electorate, Mair (2013) prominently detects a dealignment between political parties and the electorate in Western democracies. In contrast to a professionalization of political communication in MP offices and digitalization presumably closing the ranks between voters and their representatives, the notion of an alienation between the political elite and their electorate is widely shared (Merkel, 2017; Rodríguez-Pose, 2018). In fact, Merkel (2018) qualifies that it “depends strongly on the action or inaction of political elites and their interaction with the citizens whether challenges escalate into crisis.”

Zooming in on this interaction, scholars analyse politicians’ direct constituent communication. Particularly, large scale real-life experiments that consist of responsiveness tests with varying treatments have become fashionable (Costa, 2017). By and large, these studies draft an e-mail or a letter with a request or a question posed by an artificial citizen to a parliamentarian or the administration. Then, the officials’ responses are evaluated quantitatively (who answers whom how often?) and qualitatively (who answers whom how?). In an international comparison, response rates to citizens inquiries are found to be comparatively high in Germany ranging between 63 to 79 percent for Bundestag MPs (Bol et al., 2021; Heß et al., 2018) and 79 percent for local governments (Grohs et al., 2016).

A vivid debate evolved around the question whether parliamentarians acted strategically as rational vote share maximisers by answering selectively – rather responding to partisan than non-partisan groups. For mainstream politicians this translates into being more responsive to the majority society and less responsive to ethnic minorities. Whereas Butler and Broockman (2011) find that in the US politicians rather respond to a White than a Black alias, when Levine and Glick (2017) artificially contacted public officials no such discrimination is revealed – however a lower response rate is found regarding Hispanic names. Broockman (2013) adds that African American legislators in the US are more responsive to African Americans even from outside their constituency. In Germany, fake citizen requests with Turkish alias to local governments are not found to trigger an unproportionally low number of responses (Grohs et al., 2016). Although, mixed evidence is brought forward regarding ethnically motivated discrimination of citizens’ requests, elected officials’ response rates to ethnic minorities turn out to be significantly lower (Costa, 2017): Parliamentarians seem to respond selectively.

Such an empirical finding contradicts a political world absent of opportunity costs, where democratically elected emotionless politicians understand their role as one of serving their constituents and explaining political decisions truthfully, and where all requests might be answered equally and thoroughly. In fact, such a world is not ours. Political representatives and their teams face opportunity costs, time pressure, the looming risk of not being re-elected and their individual cognitive biases.

All these factors affect why as well as how representatives respond to citizens inquiries. Intending to learn more about the mechanisms behind representatives’ communication behaviour, a large body of literature focuses on the causal mechanism behind political communication and disentangles intrinsic and extrinsic motivational drivers (Bénabou and Tirole, 2003). A representative is understood to be intrinsically motivated in answering if the response is driven by their view how a dutiful politician is supposed to behave (Norris, 1997). In line with the concept of intrinsic motivation, parliamentarians are expected to not discriminate between specific groups in the electorate. In contrast, a representative is understood to be extrinsically motivated in answering if the response is primarily driven by the goal to persuade voters for the next election (Cain et al., 1984). In line with the concept of extrinsic motivation, Butler et al. (2012) compare selective answering to an election campaign’s “microtargeting”; Bol et al. (2021) reveal higher response rates of directly elected Bundestag MPs and if an inquiry includes a personal vote intention. While intrinsic motivation is generally found to be an important driver behind politicians’ communication behaviour, extrinsic motivations often seem to dominate the decision whether a request is answered or not (de Vries et al., 2016), crowding out intrinsic motivations (Giger et al., 2020).

In contrast to other studies that exploit the specific design of the political system to identify extrinsic motivation as a cause for varying response rates, the present experiment uses a topic specific treatment and focuses on the respective variation in salience for the different parties. If only intrinsic motivation mattered in answering citizens inquiries, the response rate would be equal over topics and independent of parties’ issue salience. If, however, as the literature stresses, extrinsic motivations mattered even more, citizen request should be answered in line with the party’s issue salience. A constituent who is particularly concerned about a specific issue is easier to be convinced by a party who has placed the respective issue as a core topic on their agenda.

H1: The higher the issue salience of the addressed politicians, the higher the responsiveness to respective citizen requests.

Once representatives or their staff decide to answer an inquiry, the question is, how do they deal with the respective content? In an elaborately set-up experiment Grose et al. (2015) carve out the strategical behaviour of US senators addressing their constituents. Senators are found to not explicitly lie, but tailor the truth in their answers depending on whether they face a partisan or an opposing letter on a highly politicized topic. Having said that, senators do not answer more often to letters that represent their own position as would have been expected from Larson’s (1990) assumption that politicians prefer to steer communication by setting the discussed issues and not vice versa. In contrast to Mayhew’s (1974, p. 63) past observation that “on a controversial issue a Capitol Hill office normally prepares two conflicting letters [(pro and con)] to send out to constituent letter writers”, Grose et al. (2015) detect senators’ more subtle and less contradictory but equally efficient ways to influence voters in one-to-one communication.

Again, if responsive democratic representatives followed the intrinsic idealistic conception, all requests would be answered ‘neutrally’, by correcting potentially wrong facts and supplying a political interpretation independent of the parties’ issue positioning. If extrinsic motivations mattered, however, politicians’ toleration of ‘fake news’ should depend on the party’s issue positioning. Since a fake fact could imply a certain topic was more pressing than it actually is, it might easily be tolerated following a party’s political agenda. Hence, a high ‘fake news’ toleration in line with party positioning implies a conscious, rational political behaviour intending to maximise the vote share. Such an interpretation, however, understates the cognitive constraints that bind politicians’ intrinsic and rational behaviour. Based on the theory of motivated reasoning (Kunda, 1990), recent research has progressed in understanding political behaviour psychologically and shown the power of a confirmation bias (Jakobson, 2010; Knobloch-Westerwick et al., 2015; Westerwick et al., 2017). Such a cognitive bias could also drive representatives to unconsciously ignore partisan misinformation in constituent inquiries – without being aware of tolerating fake-news (Kappes et al., 2020). Both extrinsic vote share maximisation and confirmation bias imply a higher ‘fake news’ tolerance in line with a party’s issue positioning.

H2: The more radical the issue positioning of the addressed politicians, the higher the fake-news tolerance to respective citizen requests.

3. ‘Fake news’ operationalization and research design

The fact that the interests behind the experiments outlined above is a better understanding of the elite-electorate disconnect in times of highly polarized digital communication (Mair, 2013) makes it surprising that the details of right-wing communication and manipulative ‘fake news’ remain in the dark. Populist right-wing politicians regularly claim to be the true representatives of the ‘common people’ and their ‘volonté générale’ as only they speak their language (Corner, 2017; Mudde, 2007). What is more, the rise of PRRPs is often associated with certain groups that have putatively been abandoned by the political elite – the “left-alones” who reside in “places that don’t matter” (Rodríguez-Pose, 2018). In fact, the secular decrease in turnout⁹ has in parts been reversed by the rise of PRRPs. The new political competitors have even be understood as a “corrective for democracy” (Mudde and Kalwasser, 2017),¹⁰ reinforcing the pressing question of politicians’ responsiveness to (potential) voters. Hence, a misinformation and responsiveness experiment including the right-wing putative representatives of the ‘true people’ to personal citizens inquiries bridges the gap between the different strands of literature.

The major difficulty when operationalizing ‘fake news’ in an experimental setting is hidden in the normativity of political communication. What qualifies as misinformation might not be as clearly set as critics of ‘fake news’ might suggest. In an intent to overcome the “grey zone” (Tandoc et al., 2018) between actual misinformation and ordinary tailoring of answers in an electoral campaign, the present experiment is designed to quantify the topic specific spread and toleration of misinformation by party line. Therefore, the two most controversial issues of the past elections are picked to design topic specific misinformation inquiries. Whereas the federal election of 2017 is widely interpreted as a vote on immigration to Germany (Korte, 2019) – with the AfD winning a historically high share of 12.7 percent – the election to the European parliament in 2019 was dominated by the climate change issue (Kaeding et al., 2020) – with the German Green Party skyrocketing to a 20.5 percent election result. Fictional constituent inquiries are therefore drafted around these core issues of recent political competition in Germany.¹¹ Comparable to the low-cost spread of misinformation in social media, parliamentarians are simply demanded to consent (like/ share) or disagree (dislike) with a flawed fact. In contrast to the tempting approach of picking a piece of misinformation that has already been circulated by politicians, the experiment rather employs the actual misperception of party supporters on very general and clearly verifiable facts. To generate comparable results for political extremes, an immigration vignette employs the numerical overstatement of a negative characteristic of immigrants and a climate change vignette employs the numerical understatement of the progress in fighting climate change in Germany.

More specifically, for the immigration vignette parliamentarians are asked to evaluate the misinformation whether ‘actually’ 48 percent of immigrants in Germany were unemployed in line with AfD supporters’ perception (German Economic Institute, 2020)– whereas the true value was only 14.4 percent in 2020 (Federal Employment Agency, 2021a). For the climate change vignette it is polled whether the current share of renewable energy of total electricity consumption in Germany lied ‘merely’ around 35 percent in line with German Green Party supporters’ perception (German Economic Institute, 2020) – whereas the true value increased to already 45,4 percent in 2020 (Bundesumweltamt, 2020).¹² By triggering AfD politicians to share misinformation in line with their anti-immigration issue positioning and in line with their electorate and by triggering Green politicians to share misinformation in line with their pro-environment positioning and their electorate, the present experiment is designed to reveal the truth behind the mutual misinformation recriminations. As a control cohort these requests are also sent to parliamentarians of all other parties. As a control topic the Germany-wide perceived unemployment rate of 24 percent during the current economic crisis (German Economic Institute, 2020) is sampled – whereas the official average 2020 unemployment rate was 5.9 percent (Federal Employment Agency, 2021b).

To increase responsiveness in line with Bol et al. (2021) and Giger et al. (2020), the fictional inquirer pretends to be a constituent, to sympathize with the addressed politician and to offer a potential vote for future elections. By stating to live in the parliamentarians’ constituency, the probability that the request is forwarded to and answered by a central representative from the faction is reduced. Regarding the issue, however, the inquiries are designed salient but neutral (neither pro nor con regarding the piece misinformation: “is it really true that […]”) to test politicians’ responsiveness and the spread of misinformation as generally as possible: The enquirer reveals to feel insecure regarding the conflicting information that circulated in the debate on an issue and asks for the parliamentarian’s assessment. For reasons of comparability, the empirical question refers to the federal level. Due to the inquiries’ similarities in style, each parliamentarian receives only one of the three vignettes.

To assess the factional differences of responsiveness and misinformation in electorate-parliamentarians communication, first, a database of all German national and state parliamentarians is set-up including their names, gender, e-mail addresses, the states where the constituencies are located as well as their faction, faction size, government participation by faction, issue positioning, upcoming elections in 2021 and salience by faction. Parliaments that do not offer such a list online (e.g. for data security reasons) were contacted and finally supplied all necessary information. This renders a complete mailing list of 709 national and 1870 state parliamentarians as of August 2020 – of which the six major factions that can be comprehensively analysed make up 703 national and 1800 state MPs.¹³ Second, and comparable to Bol et al. (2021), two e-mail addresses were created consisting of common German names and surnames to circumvent potential responsiveness bias in relation to gender or ethnicity (Butler and Broockman, 2011). The common German e-mail service t-online.de serves as an e-mail provider.

The e-mails were sent to parliamentarians in two waves to spread the same e-mail content over several days and thus avoid being detected as a fake inquiry or spam by the parliamentarians’ staff. A third of each faction’s parliamentarians receives either an immigration, climate change or unemployment related inquiry. The distribution of issues is randomized over parties and states.¹⁴ Time spans between the waves are kept to a minimum of one week to reduce the probability of political events that complicate comparability between the waves. To increase responsiveness, the e-mails were sent in January 2021 avoiding parliamentarian holidays and election periods. Answers are coded as valid if they are provided within a six-weeks period. In sum, three different treatment e-mails (migration, climate change, unemployment) are sent by two different aliases in two separate waves, yielding a 3 x 2 x 2 experiment set-up. This allows a robustness check to verify that the different alias or waves did not yield statistically different results regarding the treatment.

The experiment enables an analysis of the two dependent variables of interest. First, the share of parliamentarians who respond to the inquiry but remain quiet on the quantitative question or actively tolerate the piece of misinformation (’fake news’ tolerator); second, the topic-related response rate by political faction. Thus, the analysis reveals the degree of politicians’ topic specific tolerance of misinformation in Germany. Comparing the level of misinformation with parties’ issue positioning and salience analyses if correlations emerge by chance or follow a strategical rationale. Moreover, an explorative screening of the answers allows a first glimpse on the qualitative nature of narratives that are conveyed by the different parties.

Regarding Hypothesis 1 politicians of the AfD are expected to be most responsive to citizens requests touching on the immigration dimension. Since immigration is a salient topic for the Green Party as well, high response rates are also expected for representatives of the Green party. Moreover, the highest (lowest) responsiveness to citizens requests touching on the environmental dimension is expected from politicians of the Green Party (the AfD).

Regarding Hypothesis 2 it can be specified that the highest (lowest) tolerance for fake-news overstating immigrants’ burden for the German economy is expected for politicians of the AfD (the Green Party). The highest (lowest) tolerance for fake-news understating the achieved progress exiting fossil-fuel electricity consumption in Germany is expected for politicians of the Green Party (the AfD).

4. Results

4.1 Responsiveness: ‘defender of the true people’ fail to live up to their own demands

Of the 2,503 addressed parliamentarians on the national and state level 1,257 (50.2 percent) provided an answer.¹⁵ The response rate of 63.6 percent for Bundestag parliamentarians lies within the range that has been identified by similar experiments on the national level in Germany (Bol et al., 2021; Heß et al., 2018). The response rate on the state level turns out to be significantly lower (44.9 percent) raising awareness for structural differences between the different federal layers. When zooming in on the state level, evidence for extrinsically motivated response behaviour is reflected in higher response rates in the eight states with elections in 2021 (49.9 percent), in contrast to a response rate of only 40.8 percent in states without upcoming elections. In general, and despite the specificity of the requests that did not fit into most addressed parliamentarians’ issue specialization, the response rate reveals German parliamentarians’ high willingness to engage with citizens in issue specific topics. Particularly, in an international comparison (Costa, 2017) this first piece of evidence points to the fact that after all a lack of responsiveness might not be the driver of a putative dealignment between politicians and citizens in Germany.

Figure 2: Response rates to fake-news query by party, in percent; source: Own depiction

Figure 2 dives deeper into the responsiveness analysis and provides an overview over response rates by party and citizen request. First, response rates are the lowest for AfD politicians and the overall difference to the other parties is statistically significant at least on the five percent significance level.¹⁶ Only 39.2 percent of AfD parliamentarians provide an answer to the experiment’s citizens inquiry. For every of the three different citizens requests Germany’s self-declared ‘defender of the true people’ turn out to be less responsive than members of all other major political parties. Also splitting the sample by Bundestag and Länder parliaments does not alter this finding. Regarding responsiveness, the AfD rather seems to be the cause than the saviour of the putatively missing link between citizens and politicians.

Second, response rates are comparable between the unemployment (53.7 percent) and the climate change (53.8 percent) query, but significantly lower for the immigration citizen request (43 percent). Whereas the response rates’ ordering differs substantially for CDU/CSU, FDP, DIE LINKE and SPD, the Greens come second last in the unemployment and immigration treatment and only manage to leave behind FDP parliamentarians in responsiveness to the renewable energy inquiry. Fundamentally, the AfD ranges far behind the other established parties. Even when being approached by a citizen who is ‘concerned’ by immigration and shares the AfD supporters’ severe overestimation of immigrants’ unemployment rate, only 36.8 percent of the populist politicians happen to provide an answer – in contrast to a 43.4 percent response rate among non-AfD politicians.¹⁷

To test the relationship between issue salience and responsiveness, Table 1 formalizes the discussed findings by providing logistic regressions with the binary response variable as the dependent variable.¹⁸ As the independent variable of interest, the party specific salience in the issues environment, immigration and economics are added respectively from the Chapel Hill Expert Survey (2020).¹⁹ Furthermore, on the individual level, the regressions control for Bundestag affiliation and gender. On the party-state-level, the regressions control for government participation and faction size as well as for scheduled elections in 2021.²⁰ Standard errors are clustered on the state level. Hypothesis 1 follows the rationale that parliamentarians are more willing to engage in their party’s core issue. Hence, the highest response rate to the immigration request – that is based on a piece of misinformation circulation among AfD supporters – is expected from the AfD. The highest response rate to the climate change issue – that is based on a piece of misinformation circulation among Green Party supporters – is expected from the Green Party.²¹ Figure 2 reveals that this is not the case. AfD politicians rather turn out to be less responsive than parliamentarians of established parties – independent of the issue treatment. Table 1 shows that ceteris paribus no general relationship between party specific issue salience and response rates can be carved out for the overall party system. In all three citizen requests there is no statistically significant relationship between issue salience and responsiveness. In contrast, national parliamentarians are more responsive than their state level counterparts. Furthermore, an election year goes along with higher responsiveness in the climate change issue. Finally, Column 4 reveals that taken together and ceteris paribus the low responsiveness of AfD parliamentarians is not statistically significant beyond the control variables. In fact, AfD politicians are less often national than state parliamentarians, they do not form any government and their factions are smaller in size than those of other parties – characteristics that show significant positive correlations with responsiveness in bivariate regressions. Hence, the AfD’s responsiveness is the lowest, but the differences to established parties are at least partly explained by party’s peculiar characteristics.

Table 1: Logistic regression models: response rate; robust standard errors in parentheses, clustered at Länder level; *** p<0.01, ** p<0.05, * p<0.1

4.2 Numerical answers: Few outliers beyond the populist radical right

Figure 3 plots the numerical deviations of politicians’ answers from the official statistics by party averages. For this depiction, the official values are deducted from the politicians’ answers; provided links and attachments in the politicians’ answers are converted into a numerical format. If parliamentarians referred to the local unemployment rate or the share of renewable energy (despite being asked for national level) the respective numbers are evaluated. Although these ‘wrong’ answers cause a certain noise, in general, the numerical answers range in a narrow corridor around the official statistics. In fact, most parliamentarians who provide a numerical answer simply supply the official data and include the respective source. However, even when ignoring ‘fake news’ toleration, the answers by AfD politicians stand out as an extreme for every citizen inquiry. Whereas parliamentarians (who contest the ‘fake news’) of all parties on average manage to correctly specify the unemployment rate, the AfD provides a number of 6.6 percent which is ten percent higher than the official number of 5.9 percent in 2020. What is more, AfD politicians (who contest the ‘fake news’) play down the share of renewable energy (43.3 percent) that spiked to 45.4 percent of German energy consumption. And they numerically report a 23 percent too high unemployment rate of immigrants (17.8 percent) although a quick internet search reveals the official number of 14.5 percent.²²

Figure 3: Numerical deviation from official value, by party, in percent²³; source: Own depiction