To understand corporate governance in the United States, one must understand the voting behavior of mutual funds. In our paper The Party Structure of Mutual Funds, we develop the first systematic account of the structure of mutual fund preferences over corporate governance. We focus on two basic questions. First, what are the main ways in which mutual funds differ in their corporate governance preferences, as reflected in how they vote? Second, given that variation in voting behavior, what are the characteristic “types” of mutual funds in terms of their corporate governance philosophies?
To answer these questions, we use a comprehensive sample of mutual funds’ votes on 181,951 proposals from 5,774 portfolio companies by 4,656 mutual funds, covering the years 2010-2015. We apply a set of unsupervised machine learning techniques to the data to distill the key patterns in how mutual funds vote. First, we apply a type of principal components analysis to summarize each mutual funds’ voting behavior in terms of their location in a two-dimensional “preference space.” Figure 1 below plots funds’ locations. Each dot represents a mutual fund, and we also mark with triangles the average location of the funds advised by each of a set of prominent advisors.