We expand on research concerning the well-pronounced influence of geographical peer groups on human behavior. For this purpose, bank-specific risk-taking behavior and its relation to culturally close banks – measured by geographical as well as linguistic distance – is examined. We hypothesize that the level of risk taken by an distinct bank can be explained by the risk-taking behavior of other culturally close banks. Using a complete panel survey of all 1,111 separate and independent German cooperative banks from 2007 to 2010, we show with a high level of significance that banks adapt to the behavior of their culturally defined peer group. Interestingly, linguistic distance is superior to the geographical proxy. Results are robust after controlling for typical macroeconomic, bank specific, and – to eliminate unintentional herding – regional determinants. Our results are also robust to common econometrical and economic specifications. We amend existing literature on geographical herding firstly by a full census of German cooperative banks. As each cooperative bank is privileged with territorial exclusivity, our research is based on an intersect-free full coverage of the entire national territory. Secondly, we are able to refine research on geographical herding by measuring cultural vicinity via linguistic, i.e. dialect, proximity. Based on the evidence of banks selecting their peer group not by a “best-in-class” approach, but rather by dialectical proximity, we can show evidence of irrational herding resulting from psychosociological phenomena, such as mere exposure, as well as conformity effects.