What drives mutual fund asset concentration?

TL;DR

This paper investigates the concentration of assets in mutual funds, proposing a market efficiency-based model that explains the observed size distribution and asset concentration, emphasizing the slow convergence to steady-state and the dominance of random growth processes.

Contribution

It introduces a new time-dependent solution for mutual fund growth equations, demonstrating that asset concentration is primarily driven by market efficiency and random growth, with a better fit to empirical data than previous models.

Findings

The asset size distribution is well-approximated by a log-normal distribution.

Market efficiency explains the asset concentration more than transaction costs or behavioral factors.

Relaxation to the steady-state distribution is extremely slow, making the steady-state solution less relevant.

Abstract

Is the large influence that mutual funds assert on the U.S. financial system spread across many funds, or is it is concentrated in only a few? We argue that the dominant economic factor that determines this is market efficiency, which dictates that fund performance is size independent and fund growth is essentially random. The random process is characterized by entry, exit and growth. We present a new time-dependent solution for the standard equations used in the industrial organization literature and show that relaxation to the steady-state solution is extremely slow. Thus, even if these processes were stationary (which they are not), the steady-state solution, which is a very heavy-tailed power law, is not relevant. The distribution is instead well-approximated by a less heavy-tailed log-normal. We perform an empirical analysis of the growth of mutual funds, propose a new, more accurate size-dependent model, and show that it makes a good prediction of the empirically observed size distribution. While mutual funds are in many respects like other firms, market efficiency introduces effects that make their growth process distinctly different. Our work shows that a simple model based on market efficiency provides a good explanation of the concentration of assets, suggesting that other effects, such as transaction costs or the behavioral aspects of investor choice, play a smaller role.