Using data about votes emitted by funds in corporate meetings held by US banks from 2003 to 2013, we propose a novel approach based on eigenvalue decomposition to address the issue of communality in voting decisions. Our results indicate that there is a main underlying feature that contributes to explain this voting behaviour. Also, a dimensionality reduction could be accomplished so that a subset of the original data can replicate the sample. These findings confirm that there may be a sort of homogeneous or systematic component when it comes to explain the voting pattern into the banking industry.