BROKERAGE COMMISSIONS AND MUTUAL FUND PERFORMANCE

• Author: Lei Zhou,Miles Livingston
• Published date: 01 September 2015
• DOI: http://doi.org/10.1111/jfir.12060
• Date: 01 September 2015

BROKERAGE COMMISSIONS AND MUTUAL FUND PERFORMANCE

Miles Livingston

University of Florida

Lei Zhou

Northern Illinois University

Abstract

In this article we analyze fund-level data on brokerage commissions paid by diversified

U.S. equity mutual funds from 2001 to 2011. We find that brokerage commission per

dollar traded has a positive and significant impact on mutual fund performance,

indicating that funds paying premium brokerage commissions were able to improve

performance net of all expenses. The positive impact of premium brokerage services

(better execution quality, timely research reports, etc.) purchased through higher

commissions is more pronounced during volatile market times.

JEL Classification: G20, G23, G28

I. Introduction

Each year, U.S. mutual funds pay billions of dollars for brokerage commissions. Yet

previous studies on mutual fund expenses rarely investigate brokerage commissions.

We use a unique data set provided by Lipper to test empirically if and how brokerage

commissions affect fund performance.

Brokerage commission rates have experienced a secular decline over the last

three decades, partly due to the deregulation of the brokerage industry and the rise of

discount brokerage firms (French 2008). Yet the business model of full-service

brokerage firms is far from dead. Full-service brokerage firms still charge premium

commissions, or commissions significantly higher than the marginal costs of order

execution (Goldstein et al. 2009).

Proponents of the full-service brokerage model argue it is optimal to bundle

order execution with other premium brokerage services. In return for the premium

commissions, full-service brokerage firms provide premium brokerage services,

including timely research reports, better order execution, and profitable initial public

offering (IPO) allocations (Goldstein et al. 2009; Goldstein, Irvine, and Puckett 2011;

Aitken, Garvey, and Swan 1995). Brennan and Chordia (1993) demonstrate through a

We thank Lipper for providing mutual fund brokerage commissions data. We thank the referee, Jonathan

Fletcher, and the associate editor, Robert Faff, for their comments. The paper has benefited from discussions with

Xiaohui Gao; seminar participants at the University of Florida, University of New South Wales, and Victoria

University; and participants at the 2014 FMA conference.

The Journal of Financial Research Vol. XXXVIII, No. 3 Pages 283–303 Fall 2015

283

© 2015 The Southern Finance Association and the Southwestern Finance Association

RAWLS COLLEGE OF BUSINESS, TEXAS TECH UNIVERSITY

PUBLISHED FOR THE SOUTHERN AND SOUTHWESTERN

FINANCE ASSOCIATIONS BY WILEY-BLACKWELL PUBLISHING

theoretical model that (under some conditions) it is optimal and more efficient for full-

service brokerage firms to sell information services and receive payment in the form of

premium commissions. Grossman (1976) and Grossman and Stiglitz (1980) argue that

the expenses of informed investors should be offset in equilibrium by added returns,

implying a positive impact of premium brokerage commission on mutual fund net

performance.

On the other hand, premium br okerage commission might reflect the

opaqueness of brokerage commi ssions, weak fund governance, and /or the weak

competitive nature of the brokerage industry. Critics argue that brokerage

commissions are hidden costs f or fund investors because commi ssions are not

disclosed in the reported annual fund expense ratio, the fund annual report, or the fund

prospectus. Because of this op acity, mutual funds might hide ot her expenses as

premium commissions (Edelen, Evans, and Kadlec 2012). Furthermore, the ability of

some brokerage firms to charge above-order-execution-cos t commissions might be an

indication of weak competition in the brokerage industry and rent-seeking behavior of

large and dominant brokerage firms. The empirical implication of this view is that

premium brokerage commiss ions should have a negative impact on fund performance.

We use a unique data set provided by Lipper to test empirically whether

premium commissions paid by mutual funds improve or reduce fund performance by

examining the impact of mutual fund brokerage commissions on mutual fund

performance net of all expenses and costs. This data set has a large and diverse number

of mutual funds over an 11-year period.

A major innovation of the study is the development of an alternative measure of

commission: commission per dollar traded. Previous studies on commissions use

commission per dollar of assets. This measure is comparable to other fund expense

measures in the sense that they are all expressed as a percentage of total assets. However,

commission per dollar of assets does not properly measure premium commission, or

above-order-execution-cost commission, because it is driven by both the commission

rate and the total amount of trading. As a result, we develop and use a second commission

measure, commission per dollar traded, which adjusts for the level of trading activity and

better measures premium commission.

Two measures of performance are used: annual benchmark adjusted fund returns

and annual risk-adjusted four-factor alphas (Carhart 1997). We find that brokerage

commission per dollar of assets has no significant effect on fund performance.

Commission per dollar traded has a positive and significant effect on both measures of

fund performance. Commission per dollar traded adjusts for the effect of turnover and,

therefore, isolates the effect of premium brokerage commissions for individual trades.

Improved performance from premium brokerage commissions can be the result of better

execution of orders and/or the provision of valuable information by brokers to funds

paying a higher commission rate. In sharp contrast, the expense ratio and turnover ratio

have negative effects on all measures of net performance.

In addition, the effect of commission per dollar traded on fund performance

is more pronounced when stock market volatility is elevated, but nonexistent when

the market is calm. The value of premium commissions is more important during

volatile market times.

284 The Journal of Financial Research