Improving The Explanation of Electoral Behavior Through a Combination of Historical and Local Context – The Case of The AfD Result at The General Election in Germany in 2017

In a recent analysis, Homola et al. (2020) showed that the historical legacy of the Nazi Germany is still shaping the views and also the elections in Germany nowadays. Precisely, they showed that proximity to former Nazi concentration camps is associated with higher political intolerance, xenophobia and also the voting results for right-wing extremist parties. Two basic ideas can be taken from this piece: 1) historical developments might assert influence on today’s people and their voting behavior, even if these are the children, grandchildren or even great-grandchildren of the people, who have witnessed the historical time in question; 2) certain factors influencing the voting behavior can be assumed to have a strong local clustering – i.e. even if it is not completely clear which factors are the real causal determinants, it might make sense to assume elections to be strongly shaped at the local level. And this has to be reflected by the analytical approach. In my paper, I will combine these two findings in order to test to what extent a historically and spatially conscious approach can help to better understand voting dynamics in general. I will apply this approach to the analysis of the AfD result at the 2017 general election in Germany and will present two options of including local context in election studies that operate at a rather low aggregate level. In my example, I make use of a newly compiled dataset at the municipality level in Germany. First, local context will be included in a direct way, measured by distances from the municipalities to relevant geographical markers that can be associated with a long term or a short term historical legacy. Based on the works cited above, these spatial points will include concentration camps in order to test for the long term legacy of Nazi Germany. In contrast to Homola et al. I will not only test for the effects of the main concentration camps but also include the relevant subcamps. Furthermore, the distances to first reception facilities for refugees during the so called “refugee crisis” (2015-16) will be used in order to test for the shorter term historical legacy of this incisive development in the recent past. Additionally, I will control for the distance to the Eastern border, as this variable has proven to be relevant in the context of earlier studies (Jäckle et al., 2019). Second, I introduce the method of Geographically Weighted Regression (GWR) as an option to answer the more localized questions by actively modeling spatial non-stationarity. The resulting coefficient-maps will show indeed that most effects (from modernization loser theory to the contact hypothesis) vary significantly and considerably in terms of strength throughout Germany, rendering the interpretation of any global effects very difficult.