<i>CFA Institute Journal Review</i> (formerly <i>CFA Digest</i>) summarizes "How Rational and Competitive Is the Market for Mutual Funds?," by Markus Leippold and Roger Rueegg, published in <i>Review of Finance</i>, June 2019.
Actively managed funds generally cannot generate alpha relative to index funds. Compared with index funds, active funds have a small-cap bias and negative sensitivity to changes in the VIX. Past losers and small and high-fee funds tend to have negative alphas. Evidence shows that US funds have diseconomies of scale in generating alpha.
What Is the Investment Issue?
Active investment proponents believe that even in rational markets, active managers can still generate alpha. According to basic economic principles, they believe, agents earn economic rents; investors are willing to pay higher management fees to reward more skillful active managers. In this case, the alpha before fees can be positive, but because the market is competitive, investors will compete with each other for the winning funds and push up the fees, eventually dragging the alpha down to zero after fees. According to the efficient market hypothesis, the market price reflects all price-sensitive information, so it is not possible to buy undervalued stocks. Followers of passive investment argue that active funds are a zero-sum game before fees; positive alphas are offset by negative alphas. After charging fees, alphas will turn negative.
The authors investigate whether active fund managers generate alphas, both before and after fees. Then, they examine the factors that explain the performance differences between active and index funds. Building on the findings of previous researchers, the authors further explore whether fund size, past performance, and fees affect the size of alphas.
How Did the Authors Conduct This Research?
The authors empirically examine the mutual fund sample from the Morningstar database for December 1991–December 2016. The sample consists of 61,269 funds and includes both active and passive funds. Based on their characteristics—retail or institutional, equity or fixed income, from the United States or other regions, and so on—the funds end up divided into 63 categories.
Because the data are prone to be affected by serial and cross-sectional dependencies, the authors propose a robust alpha test based on a Studentized block bootstrap, which enhances the accuracy of an inference for dependent time-series data.
To test whether active investing is a zero-sum game, the authors measure the performance of value-weighted portfolios in each category and then compare it with an investible benchmark through a robustness test.
After quantifying the performance difference between active and passive investments, the authors regress the performance difference against the three-factor model and changes in the CBOE Volatility Index (VIX) to explore why active funds performed differently from passive funds.
The authors investigate whether past winners, fee amounts, and diseconomies of scale explain the alphas of active funds.
What Are the Findings and Implications for Investors and Investment Professionals?
Active investing is a zero-sum game after fees; only 3 categories (out of 63) significantly underperformed the benchmark, whereas the other 60 have insignificant alphas compared with their corresponding benchmarks. Before fees, several categories significantly outperformed the benchmark, but all are insignificant after fees.
US funds in general fail to extract value from the market. Both retail equity and fixed-income funds added almost zero value for investors after fees. Even worse, the value extracted by active US institutional funds was less than the fee charged. The inferior performance of US funds could possibly be explained by the conjecture that greater competition in the mature and large mutual fund industry in the United States leads to diseconomies of scale, making it difficult for fund managers to beat the benchmark.
The authors find that 52% of US funds were excessively overfunded, far exceeding the 32% of overfunded European funds, showing that the inferior performance could be attributed to an oversized US active fund industry.
To explain the performance difference between active and index funds, the authors observe that active managers prefer growth stocks over value stocks, which contradicts what is proposed by the three-factor model. It is also surprising that active funds performed worse when there were changes in the VIX. Because hedging can be costly, the authors believe that active managers prefer to make small gains by selling insurance, instead of hedging risk, during generally volatile periods.
The authors also observe that large funds, persistent winners, and low-fee funds generally perform better. Although such funds may not be able to generate significant positive alphas, investing in the opposite types of funds (small funds, losers in the previous period, and high-fee funds) certainly could lead to significant negative alphas.