We offer a new perspective on the low-beta anomaly by acknowledging the omitted-variable problem in the correlation component of beta: Correlation is "plagued" by firm size (the omitted variable) to ...