A word of advice to Angela from Theresa, Hillary and Hannelore: Campaigns do matter

We have seen more than surprising outcomes in recent elections, nationally and internationally: the UK's recent election, two German state elections (North Rhine-Westphalia and Schleswig-Holstein) and, of course, the US presidential race in 2016. What do they all have in common? Disastrous campaigns by the incumbents and incumbent parties.  

Campaign research has been a growing discipline in the social sciences for the last 25 years, bringing together electoral research, party research and communication and media studies. The field is flourishing because campaigns have become increasingly important – campaigns matter, depending on how they are conducted. Our research shows that changes in voter behaviour in recent years mean candidates have to change the way they approach campaigns. There are a number of things politicians can do to run better campaigns – and recent election results offer good evidence that many have not yet taken this message on board.

What influences voting behaviour? A well-known model developed by researchers at the University of Michigan, shows that party identification is a long-term factor, while candidates and policy issues are short-term factors. Party identification is decreasing in all established democracies. In Germany, party identification has been declining slowly but continuously over the last 30 years: from 80% in the 1980s to almost 60% in 2012[1]. The number of party members has also continuously decreased.[2]  In the US, based on 2014 data, 39% of the voters identify as independents - this is the highest percentage of independents in more than 75 years of public opinion polling[3].

Interestingly, party identification has never been very high in younger democracies, such as those in Eastern Europe. In addition, the number of so-called late-deciders - people who make up their minds just two weeks or less before an election - has risen considerably. Polls showed that in the last US election, 13% of voters waited until the final week of the presidential race to make up their minds between the candidates.

So what does that mean? If party identification wanes, policy issues and candidates become more important. Since these two short-term factors vary from election to election, they need to be communicated via campaigns. As a result, the importance of campaigns rises. And they have to be conducted in the right way. This means focusing on some specific aspects: spinning narratives and presenting a message of authenticity, using data in a well-targeted manner, and being prepared to play defense on unexpected events.

The failed campaigns of recent weeks and months offer some ideas on “right” and “wrong” campaigns.

1). Narratives matter. They are decisive. “Stronger together” wasn’t. But “Make America great again” was. As candidate for US president, Donald Trump managed to hit the bull’s eye by making millions of Americans feel patriotic and giving them the feeling of taking their needs and desired serious; French presidential candidate Emmanuel Macron appealed to voters by calling for mobilisation and optimism after years of stagnation with Hollande and also Sarkozy. Both men captured the emotional state of many voters rather than talking specifically about policies. UK Prime Minister Theresa May, however, didn’t have a convincing narrative. The only message she seemed to convey was one of overconfidence. She seemed so sure about a landslide victory that political content hardly played a role – which did not help her result. The German Social Democrat candidate to lead North Rhine-Westphalia, Hannelore Kraft, ran a campaign focused too much on her persona, and didn’t pay enough attention to voters’ dissatisfaction with the economic, social and political situation in the state.

2) Authenticity. Don’t overplay. Don’t over-script. Don’t over-professionalise. Be real. Here, again, Donald Trump – against all odds, managed to convince voters that he was an outsider, even though he was part of the elite for decades – and managed the come across as authentic. Hillary Clinton on the other hand did not. She seemed over-controlled. Unauthentic. Also, Theresa May mostly failed, see #fieldofwheat.

3). Data are important, but they are a tool. Not more, not less. You can micro-target your audience, you can pinpoint the swing voters, but you need something to talk about. Data tools can serve as a catalyst for political messages, but they cannot create them (so, back to point 1).

4) Mobilize! It may seem a bit old-fashioned, but getting to know your constituents, talking to people and convincing them to vote for a certain candidate is vital; At the end of the day, every individual counts. Data (see point 3.) can be of great use in this process, but without face-to-face campaigning, it is hard to make an authentic (point 2.) campaign based on narratives (point 1.). The successful candidates used different media, different campaign styles and different strategies, but they all had a clear vision of whom they wanted to talk to and what they wanted to say to them. Moreover, a lot of effort was put into de-mobilisation: discouraging voters who might vote for opponents. One should be aware, however, that this raises serious ethical – and in some cases even legal – questions.

5) Prepare for unexpected circumstances. Surprise can happen anytime and can quickly spin out of control. Not only Hillary’s e-mails in the US can turn a result around, but also terror attacks in the UK, badly handled interviews in Schleswig-Holstein or even natural disasters such as Hurricane Katrina or the Oder flooding.

Besides these factors that apply to all election campaigns, there is one programmatical issue that applies especially to the left. It has to recalibrate its focus: from postmaterialism back to the socio-political cleavage. Many election results show this need for a policy turn on the left, the necessity of a proper political alternative. If the left remains in paralysis, more and more voters will turn to the far right, which offers simple rhetoric on complex topics. However, such a turn to the left is nowhere to be seen in Germany, which is why forecasting tools like the new model by Prof. Mark Kayser and Arndt Leininger of the Hertie School predict the Social Democrats won’t achieve a good result in Germany’s September federal election. Even though the incumbent conservative CDU is seen losing around 6%, it will still remain the strongest party. Winners will be the far-right AfD and the Liberal Democrats (FDP).

Let us wait and see whether we see a game-changer-campaign unfolding over the summer.

[1] http://www.kai-arzheimer.com/party-identification-in-germany-not-dead-yet-1/  & https://lirias.kuleuven.be/bitstream/123456789/343067/1/MPSA+2012+Party+Identity+in+Germany.pdf

[2] https://de.statista.com/statistik/daten/studie/1214/umfrage/mitgliederentwicklung-der-spd-seit-1978/ & https://de.statista.com/statistik/daten/studie/1215/umfrage/mitgliederentwicklung-der-cdu-seit-1978/

[3] http://www.people-press.org/interactives/party-id-trend/