INVESTMENT FEES, NET RETURNS, AND CONFLICT OF INTEREST IN 401(K) PLANS

• DOI: http://doi.org/10.1111/jfir.12087
• Author: Sabuhi H. Sardarli,Thomas W. Doellman
• Date: 01 March 2016
• Published date: 01 March 2016

INVESTMENT FEES, NET RETURNS, AND CONFLICT OF INTEREST IN

401(K) PLANS

Thomas W. Doellman

Saint Louis University

Sabuhi H. Sardarli

Kansas State University

Abstract

Using a proprietary database of predominantly small to mid-size 401(k) plans

administered by nearly 400 unique third-party plan administrators (TPAs), we examine

the potential effects of a conflict of interest that arises from the TPA’s incentive to

recommend its funds for the investment menu. We find that investment fees are highest

and net returns are lowest for plans administered by asset management advisory firms,

commercial banks, and insurance companies. The higher fees and lower returns are

related to the existence of TPA proprietary funds in these plans’menus and to proxies for

agency conflicts capturing proprietary trading and revenue sharing.

JEL Classification: G10, G11, G20, G21, G23

I. Introduction

By the end of the year 2014, $6.8 trillion was invested in defined contribution (DC) plans,

up from $3.7 trillion in 2005, and compared to $1.7 trillion in 1995 (Investment Company

Institute 2015). Given the growing importance of employer-sponsored DC plans to

retirement savings, understanding the various forces and conflicts that affect savings

accumulation efforts in these plans is a topic of great interest to government agencies,

academics, and practitioners. Clearly, individual and employer contributions matter, but

so do fees and net performance of the investment funds in the plan menu. For this reason,

the actions of third-party plan administrators (TPAs) are under increasing scrutiny. A

TPA is a financial institution hired by the plan sponsor to provide the administrative

functions of a 401(k) plan. Often, TPAs are also relied on to provide expertise when

choosing the plan’s investment menu. This reliance on TPAs by plan sponsors creates the

potential for agency conflicts that can have important implications for plan investment

We would like to thank those at BrightScope Inc. for their invaluable help with this project. Without sharing

with us their unique data set and their expertise in the industry, this project would not have been possible. We would

also like to thank Michael Alderson, Naresh Bansal, Brian Betker, Mark Flannery, Melissa Frye (associate editor),

Joel Houston, Chris James, Miles Livingston, Andy Naranjo, Tareque Nasser, Jay Ritter, Michael Ryngaert, Olgun

Sahin, and the anonymous referee, as well as seminar participants at the Kansas State University, Saint Louis

University, Texas Tech University, University of Florida, University of Tulsa, and the Financial Management

Association annual meeting held in Chicago (2013) for helpful comments. All errors within are solely those of the

authors.

The Journal of Financial Research Vol. XXXIX, No. 1 Pages 5–33 Spring 2016

5

© 2016 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

fees and net return performance. In this article, we study the effect of identifiable agency

conflicts on plan menu investment quality across the various types of institutions acting

as TPAs in the 401(k) industry.

The ERISA fiduciary standard requires that a financial advisor’s investment

advice be based solely on the interest of the client and provides investors legal recourse

when investment advice fails to meet the fiduciary duty standards. In contrast,

nonfiduciaries can provide investment advice based on financial incentives (i.e., conflicts

of interest) as long as the advice meets the less restrictive suitability standard. Thus, it is

possible that two advisors, one serving as a fiduciary and the other not, could provide the

same conflicted investment advice to a client with only the fiduciary facing the prospect

of litigation. However, clients of fiduciaries and nonfiduciaries are equally affected by

conflicted investment advice.

A recent U.S. Government Accountability Office (2011) report provides

anecdotal evidence that TPAs often do not act as fiduciaries, naturally fueling a

long-standing concern over the potential effect of conflicted investment advice by TPAs

on retirement savings in the 401(k) industry. In February 2015, the Obama

Administration announced its support for a proposal by the Department of Labor that

would require financial institutions serving DC plan participants to meet the fiduciary

standard.

1

The underlying assumption is that participants will receive better investment

advice, or at least stronger legal recourse in the event of wrongdoing, if the TPA is a

fiduciary. The purpose of this article is to shed light on whether identifiable TPA conflicts

of interest are in fact associated with plan menu investment quality.

Despite its importance, there is limited academic research addressing this topic,

largely because of a lack of data. Many prior academic studies collect 401(k) plan data

from the 11-k filings of large publicly traded companies. These plan sponsors

predominantly hire large mutual fund families as the TPA (Pool, Sialm, and Stefanescu

forthcoming). In contrast, our proprietary database consists of 6,809 401(k) plans that

better represents the average plan sponsor and the various types of companies operating

as TPAs in the 401(k) industry. Specifically, our sample includes predominantly small to

mid-sized plans of both publicly traded and private companies, and the plans are

administered by close to 400 unique TPA companies. This unique database is crucial for

allowing a robust analysis of the important research question addressed here.

The source of agency conflict we study is TPA proprietary investment funds.

When a TPA considers investment options for a plan’s menu, naturally the valuable plan

asset flows and associated fund management fees create an incentive to include its

proprietary investment funds. This conflict of interest has the potential to affect plan

investment fees and net return performance if funds similar to the TPA’s proprietary

funds are available through other financial institutions at a lower cost. To capture the

potential effect of this conflict, we measure the proportion of a plan’s assets that is

invested in the proprietary funds of the TPA and determine whether this variable is

associated with plan menu investment quality. As an alternative, broader measure of

1

D. Michaels and A. G. Keane, “Obama Backs Tougher Rules for Brokers on Retirement Funds,”

Bloomberg.com (February 23, 2015), http://www.bloomberg.com/politics/articles/2015-02-23/obama-to-lead-

push-to-toughen-broker-rules-for-retirement-funds

6 The Journal of Financial Research