Alphabeticity Bias in 401(k) Investing

• Date: 01 November 2019
• Author: Jennifer Itzkowitz,Jesse Itzkowitz,Thomas W. Doellman,Sabuhi H. Sardarli
• Published date: 01 November 2019
• DOI: http://doi.org/10.1111/fire.12188

The Financial Review 54 (2019) 643–677

Alphabeticity Bias in 401(k) Investing

Thomas W. Doellman

Saint Louis University

Jennifer Itzkowitz

Seton Hall University

Jesse Itzkowitz

Ipsos Behavioral Science Center

Sabuhi H. Sardarli∗

Kansas State University

Abstract

Structural factors that cause irrational investmentin defined contribution savings plans are

of great concern. Using a proprietary database of 401(k) plans we show that alphabeticity—the

order that fund names appear when listed in alphabetical order—significantly biases partici-

pants’ investment allocation decisions. While we show a larger impact as the number of funds

in the plan increases, this bias is strong even when relativelyfew funds are available in the plan

menu. Importantly, our findings suggest that a more strategicordering of funds could result in

favorable outcomes for participants.

∗Corresponding author: College of Business Administration, Kansas State University, 2097 BB,

1301 Lovers Lane, Manhattan, KS 66506; Phone: (785) 532-5771; Fax: (785) 532-6822; E-mail:

ssardarli@ksu.edu.

We would liketo thank the editor, two anonymous referees, conference participants at the 2017 Financial

Management Association Annual Meeting and the 2017 Midwest Finance Association Annual Meeting,

and seminar participants at Saint Louis University and Seton Hall University for helpful comments. This

paper was partly completed during Jennifer Itzkowitz’sVisiting Scholar position at New York University

while on a Sabbatical from Seton Hall University.

C2019 The Eastern Finance Association 643

644 T. W. Doellman et al./The Financial Review 54 (2019) 643–677

Keywords: alphabeticity, 401(k) plans, portfolio selection

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

1. Introduction

Investments in 401(k) and other defined contribution (DC) plans constitute a

significant proportion of Americans’ retirement savings. According to the Investment

Company Institute (2017), as of year-end 2016, $7.6 trillion in assets were held in

DC plans, accounting for 28% of all retirement assets in the United States. As DC

plans have grownin prevalence, so have individuals’ responsibility overtheir financial

futures. Thus, it is critical to understand the factors that shape individuals’ investment

decisions in these plans. Prior work shows that several factors can preclude DC plan

participants from investing rationally: bias, behavioral inertia, framing, and a lack of

financial literacy.1We consider a new behavioral bias—alphabeticity—in the context

of 401(k) investing and examine its potential effects on participants’ investment

decisions.

Alphabeticity bias is the phenomena in which early alphabet options are chosen

more frequently than others. Twopsychological processes contribute to alphabeticity

bias: status quo bias and satisficing. Investment options within 401(k) and other DC

plans are often listed in alphabetical order. While many investors can access their

plans online and may be able to reorder their list of fund choices, individuals generally

rely on the default (status quo) list given to them (Samuelson and Zechkauser, 1988;

Kahneman, Knetsch and Thaler, 1991). Thus, because funds are initially listed in

alphabetical order, they remain in alphabetical order when participants make fund

allocation decisions within the plan.

Reliance on the status quo interacts with individuals’ tendency to satisfice re-

sulting in alphabeticity bias. When choosing between multiple alternatives, each pos-

sessing different attributes, individuals typically satisfice, where their search ceases

after the first “acceptable” option is found, even if continued searching could yield

a better result (Simon, 1956; Payne, 1976; Caplin, Dean and Martin, 2011). Thus,

when a participant searches through her plan’s menu of investment options, she may

be more likely to choose the funds appearing toward the beginning of the list. Since

401(k) fund choices with early alphabet names appear at the beginning of the list,

they will be chosen more often than later alphabet-named funds.

1A fewexamples of this literature include Agnew, Balduzzi and Sunden (2003), Benartzi and Thaler (2001,

2007), Duflo and Saez (2002), Finke, Howe and Huston (2017), Huberman and Jiang (2006), Lusardi and

Mitchell (2007), and Madrian and Shea (2001).

T. W. Doellman et al./The Financial Review 54 (2019) 643–677 645

While alphabeticity bias is shown to affect other types of decision making,

including stock choice,2differences between stock market investing and investing in

401(k) plans may mitigate the likelihood of alphabeticity bias affecting investment

choice in 401(k) plans. Stock market investors have thousands of investment options

to choose from, each with an overwhelming amount of information available for

investors to sift through, making it impossible to consider every option. In contrast,

401(k) investors typically choose from a relatively small number of funds making it

more manageable for investors to consider every option.3Because alphabeticity bias

relies on investors’ truncated search, the bias may not affect fund choice in 401(k)

plans like it does stock choice.

We find that alphabeticity bias indeed affects investment allocation decisions

in DC plans. Consider a 401(k) plan that offers the following 13 funds options

(which are listed in alphabetical order in the plan offering): Brandywine, Columbia

Acorn, Dodge & Cox Stock, Fidelity Contrafund, Fidelity Freedom Income, Fidelity

Puritan, Forward Small To Mid Cap, Harbor Capital Appreciation, Pimco All Asset

All Authority, Pimco Commodity Real Return Strategy, Royce Pennsylvania Mutual,

T.Rowe Price Blue Chip Growth, Vanguard 500 Index. We find that if the Royce fund

changed its name to American Royce Pennsylvania Mutual (a 10-position increase,

moving it to the top position when listed in alphabetical order), investment in the

fund would increase roughly by 20%, all else equal. This represents an additional

allocation of $653,000 to the Royce fund in the average plan with $32.5 million in

assets.

Next, we recognize that a fund’s position on the plan menu is more important

than a fund’s relative location in the alphabet. For example, if a 401(k) plan offers

only Vanguard funds (each starting with the wordVanguard), then the Vanguard 500

Index fund is likely to be listed first. However, if a 401(k) plan offers Vanguard funds

as well as funds from other families, then the Vanguard 500 Index fund is likely to

appear toward the end of the list. Weexploit the unique size and cross-sectional nature

of our data set to examine this more closely. We find that the same fund appearing

in multiple plans in the sample receives a significantly higher allocation when it is

listed closer to the top of the plan menu. This result offers further support of our

main hypothesis that the alphabeticity bias significantly affects investmentallocation

decisions in 401(k) plans.

2Research in other disciplines shows that alphabeticity influences many different decision processes.

Politicians with early alphabet names are more likely to be elected than their competitors (Edwards, 2015).

Scholars with last names that begin early in the alphabet are invited to review papers more often than

those with last names that start with letters later in the alphabet (Richardson, 2010). And, because they

are solicited more often, alumni with early alphabet names donate more than those with later alphabet

names (Meer and Rosen, 2011). Stocks beginning with early alphabet letters have been found to have

higher turnover, liquidity, and valuation than later alphabet stocks (Itzkowitz, Itzkowitz and Rothbort,

2015; Itzkowitz and Itzkowitz, 2016; Jacobs and Hillert, 2016).

3Plans in our sample contain an average of 20 funds. The plan with the most funds has 129 options.