Press Release of February 21, 2018
DIW Berlin study analyzes the correlation between the AfD's vote performance and different economic and sociodemographic variables at an electoral district level – The AfD performed well in western German electoral districts where there are many employees in the manufacturing industry and where incomes are low – In the eastern districts they performed better where there is a large share of elderly residents and a high density of craft businesses
DIW Berlin president Marcel Fratzscher and co-authors Christian Franz and Alexander Kritikos analyzed the correlation between the Alternative for Germany’s (Alternative für Deutschland, AfD) performance in the last parliamentary election and seven economic and sociodemografic structural variables. Certain characteristics, such as the unemployment rate or share of non-German citizens in each electoral district, appear to barely play a role. There are positive correlations for other factors, meaning that these characteristics are especially pronounced in electoral districts where the AfD performed well. The patterns are different in the eastern and western parts of Germany.
The study used a multivariate regression analysis. Seven variables were used, each available at the level of the individual electoral districts: (1) the average disposable household income, (2) the unemployment rate, (3) the proportion of employees in the manufacturing industry, (4) the density of craft businesses measured by the number of craft businesses per 1,000 inhabitants as an indicator of a the economic structure and of the district's population density, (5) the share of the population over 60 years old, (6) the share of the population with non-German citizenship (however, the data is from 2015, and significant changes have occurred since then in certain electoral districts), and (7) the high school graduation rate, which serves as an indicator of the electoral district's level of education. Source for this aggregated data is the Federal Returning Officer.
Berlin was excluded from the analysis because it is comprised of 12 very different electoral districts, but data was only available for the entire city.
“In the west, the AfD has strong support in electoral districts where the disposable household income is under the national average and the share of employees in the manufacturing industry is disproportionately high. In the east, the AfD performed more strongly in regions with an above-average share of inhabitants over 60 as well as in regions with a high density of craft businesses,” said Alexander Kritikos, summarizing the most important results. The density of craft businesses indicates the economic structure of an electoral district as well as its population: areas with a disproportionate amount of craft businesses tend to be more sparsely populated. Thus, the AfD performed strongly in rural electoral districts with large aging populations.
The study shows that single reason explanations for the AfD’s success fall short. “The AfD is not simply the party of the unemployed, those with low incomes, or east Germans—the reality is much more complex,” said Christian Franz. “Depending on whether you look at eastern or western Germany, other factors seem to be at work.” At the same time, he emphasized that the “study was concerned with structures and not with individual voting decisions and the motivations behind them."
The factors analyzed in some of the electoral districts were less able to account for the AfD’s election performance. The model underestimated support for the AfD in border regions in the eastern parts of Saxony and Bavaria in particular. The authors used the performance in the 2013 parliamentary elections of the right-wing National Democratic Party (Nationaldemokratische Partei Deutschlands, NPD) as an additional variable to map the outliers in the AfD’s proportional vote performance. There is actually a limited positive correlation between support for the NPD four years ago and high numbers of votes for the AfD in 2017. This correlation is more pronounced in eastern Germany than in the western part of the country.
“When certain structural characteristics are especially pronounced, support for established parties decreases and support for the right-wing populist party AfD rises. These characteristics include low incomes, a large manufacturing industry—fear of job loss due to automation and digitalization could play a role here—and an overaged population in rural areas,” concluded Marcel Fratzscher. “Politicians must pay special attention to those parts in eastern Germany deserted by young people. Social participation must be improved and more emphasis placed on public infrastructure to maintain basic services and make these regions attractive places to live again.”