Unwanted foreign influence? Analysis of the Norwegian parliamentary elections 2021

This study aims to investigate whether the Norwegian Parliamentary election 2021 was subject to malign information influence by foreign actors. “Information influence” is defined in this study as the spread of (dis)information and/or manipulation on Norwegian websites and in social media with the presumed intent of influencing the election results, voter participation or trust in the election itself. The Election Day was 13 September, and the investigation focuses on the periode between 1 August – 16 September 2021.

The study was conducted by the Norwegian Defence Research Establishment (FFI), in cooperation with the Scandinavian analysis agencies Analyse & Tall and Common Consultancy, on behalf of the Norwegian Ministry of Local Government and Modernisation.

Our study is based on the hypothesis that malign information influence by foreign actors has not taken place. By using different quantitative and qualitative methods, we have attempted to falsify this hypothesis. We have approached this task by mapping and analyzing the spread of (dis)information and propaganda from foreign actors on Norwegian websites, on Facebook and on Twitter. Furthermore, we have investigated inauthentic activity on Facebook and Twitter in order to identify targeted attempts at manipulation. Our study has focused on Facebook and Twitter because a substantial part of the Norwegian public debate takes place on these platforms, and because these platforms have previously been used in influence operations targeting other nations. Furthermore, these platforms provide access to data that makes it possible to conduct this kind of investigation. Foreign influence may take place on several other platforms. However, these are not included due to technical, legal and/or resource limitations. Our findings and conclusions are thus limited to the datasets investigated during the set time period. However, we consider it unlikely that a foreign actor will be able achieve a sufficient effect on the Parliamentary election without leaving traces of influence activities on Facebook, Twitter or Norwegian websites.

Through four in-depth analyses, we have not made any findings that suggest foreign actors attempted to influence the election results, voter participation or trust in the election itself. However, our analyses show that foreign, non-state actors are actively involved in attempts to spread disinformation to a Norwegian target audience, albeit with limited reach. We have also uncovered two clusters of seemingly foreign, inauthentic Twitter profiles spreading content to a Norwegian audience. Facebook and Twitter are still exploitable with regards to conducting influence campaigns, despite the platforms’ attempts to stop inauthentic behaviour and the spread of disinformation.

Our literature review shows that malign foreign influence directed at democracies is ongoing at a significant level, and is often conducted with low intensity over time. Based on the sheer scope, we consider it likely that foreign actors, with sufficient will and capability, may direct influence operations at the Norwegian population or Norwegian interests in a situation where the actor finds this opportune. Consequently, we recommend that future analyses of malign foreign influence aimed at elections should be expanded to include long-term influence directed at specific political issues and attempts to weaken people’s trust in democratic institutions.